WebKit-https.git
4 years agoRenderImageResourceStyleImage::image() should return the nullImage() if the image...
commit-queue@webkit.org [Fri, 4 Aug 2017 20:42:43 +0000 (20:42 +0000)]
RenderImageResourceStyleImage::image() should return the nullImage() if the image is not available
https://bugs.webkit.org/show_bug.cgi?id=174874
<rdar://problem/33530130>

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2017-08-04
Reviewed by Simon Fraser.

Source/WebCore:

If an <img> element has a non-CachedImage content data, e.g. -webkit-named-image,
RenderImageResourceStyleImage will be created and  attached to the RenderImage.
RenderImageResourceStyleImage::m_cachedImage will be set to null at the
beginning because the m_styleImage->isCachedImage() is false in this case.
When ImageLoader finishes loading the url of the src attribute,
RenderImageResource::setCachedImage() will be called to set m_cachedImage.

A crash will happen when the RenderImage is destroyed. Destroying the
RenderImage calls RenderImageResourceStyleImage::shutdown() which checks
m_cachedImage and finds it not null, so it calls RenderImageResourceStyleImage::image()
which ends up calling CSSNamedImageValue::image() which returns a null pointer
because the size is empty. RenderImageResourceStyleImage::shutdown() calls
image()->stopAnimation() without checking the return value of image().

Another crash will happen later when deleting the CachedImage from the memory
cache if CachedImage::canDestroyDecodedData() is called because the client
it gets from m_clients is a freed pointer. This happens because RenderImageResourceStyleImage
has m_styleImage of type StyleGeneratedImage but its m_cachedImage is set
by RenderImageResource::setCachedImage(). When RenderImageResourceStyleImage::shutdown()
is called, it calls  StyleGeneratedImage::removeClient() which does not
know anything about RenderImageResourceStyleImage::m_cachedImage. So we
end up having a freed pointer in the m_clients of the CachedImage.

Test: fast/images/image-element-image-content-data.html

* rendering/RenderImageResourceStyleImage.cpp:
(WebCore::RenderImageResourceStyleImage::shutdown):  Revert back the changes
of r208511 in this function. Add a call to image()->stopAnimation() without
checking the return of image() since it will return the nullImage() if
the image not available. There is no need to check m_cachedImage before
calling image() because image() does not check or access m_cachedImage.

If m_styleImage is not a CachedStyleImage but m_cachedImage is not null,
we need to remove m_renderer from the set of the clients of this m_cachedImage.

(WebCore::RenderImageResourceStyleImage::image const): The base class method
RenderImageResource::image() returns the nullImage() if the image not
available. This is because CachedImage::imageForRenderer() returns
the nullImage() if the image is not available; see CachedImage.h. We should
do the same for the derived class for consistency.

LayoutTests:

* fast/images/image-element-image-content-data-expected.txt: Added.
* fast/images/image-element-image-content-data.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220289 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUse MPAVRoutingController instead of deprecated versions.
commit-queue@webkit.org [Fri, 4 Aug 2017 20:07:20 +0000 (20:07 +0000)]
Use MPAVRoutingController instead of deprecated versions.
https://bugs.webkit.org/show_bug.cgi?id=175063
Source/WebCore:

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-04
Reviewed by Tim Horton.

No new tests because no behavior change. This uses a different platform class to present
an interface.

Remove deprecated MPAudioVideoRoutingPopoverController and MPAVRoutingSheet
Add MPMediaControlsViewController.

* platform/spi/ios/MediaPlayerSPI.h:

Source/WebKit:

rdar://problem/33301230

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-04
Reviewed by Tim Horton.

Remove dependence on deprecated classes MPAVRoutingSheet and MPAudioVideoRoutingPopoverController.

* UIProcess/ios/forms/WKAirPlayRoutePicker.h:
* UIProcess/ios/forms/WKAirPlayRoutePicker.mm:
(-[WKAirPlayRoutePicker dealloc]):
(-[WKAirPlayRoutePicker show:fromRect:]):
(-[WKAirPlayRoutePicker popoverControllerDidDismissPopover:]): Deleted.
(-[WKAirPlayRoutePicker _presentAirPlayPopoverAnimated:fromRect:]): Deleted.
(-[WKAirPlayRoutePicker _windowWillRotate:]): Deleted.
(-[WKAirPlayRoutePicker _windowDidRotate:]): Deleted.
(-[WKAirPlayRoutePicker _dismissAirPlayRoutePickerIPad]): Deleted.
(-[WKAirPlayRoutePicker showAirPlayPickerIPad:fromRect:]): Deleted.
(-[WKAirPlayRoutePicker showAirPlayPickerIPhone:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220288 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoDashboard bubbles sometimes show failure count instead of crash count
aakash_jain@apple.com [Fri, 4 Aug 2017 20:01:16 +0000 (20:01 +0000)]
Dashboard bubbles sometimes show failure count instead of crash count
https://bugs.webkit.org/show_bug.cgi?id=175157
<rdar://problem/33709009>

Reviewed by Alexey Proskuryakov.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BuildbotTestResults.js:
(BuildbotTestResults.prototype.resultSummarizer): Use the regex for the precise match first.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220287 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoviewport-fit changes during animated resize can cause layout size to get stuck
timothy_horton@apple.com [Fri, 4 Aug 2017 19:50:23 +0000 (19:50 +0000)]
viewport-fit changes during animated resize can cause layout size to get stuck
https://bugs.webkit.org/show_bug.cgi?id=175169
<rdar://problem/33684697>

Reviewed by Simon Fraser.

If we get a commit that changes viewport-fit state while an animated
resize is underway, and that change causes the client to push a new
minimumLayoutSizeOverride, the new size will be lost forever, and we
will get stuck laying out at the wrong size.

This is because we unconditionally apply avoidsUnsafeArea for every commit,
while most other changes that come in from a commit are ignored if we're
inside animated resize. To fix, also ignore avoidsUnsafeArea changes during
animated resize, by moving the code that keeps track of it into WKWebView
like all of the rest, and read it out of the commit in didCommitLayerTree
*after* the animated-resize early-return.

Also, fix the associated layout tests by adding a missing assignment
in _computedContentInset, which was broken in r220138.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):
(-[WKWebView _setHasCustomContentView:loadedMIMEType:]):
(-[WKWebView _processDidExit]):
(-[WKWebView _didCommitLayerTree:]):
(-[WKWebView _setAvoidsUnsafeArea:]):
(-[WKWebView _safeAreaShouldAffectObscuredInsets]):
(-[WKWebView _didChangeAvoidsUnsafeArea:]): Deleted.
* UIProcess/API/Cocoa/WKWebViewInternal.h:
* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::resetState):
(WebKit::WebPageProxy::setAvoidsUnsafeArea): Deleted.
* UIProcess/WebPageProxy.h:
(WebKit::WebPageProxy::avoidsUnsafeArea const): Deleted.
* UIProcess/ios/PageClientImplIOS.h:
* UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::didChangeAvoidsUnsafeArea): Deleted.
* UIProcess/ios/WebPageProxyIOS.mm:
(WebKit::WebPageProxy::didCommitLayerTree):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220286 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRebaslining fast/text/font-selection-font-loading-api-parse.html for iOS 11.
jlewis3@apple.com [Fri, 4 Aug 2017 19:44:04 +0000 (19:44 +0000)]
Rebaslining fast/text/font-selection-font-loading-api-parse.html for iOS 11.

Unreviewed test gardening.

* platform/ios-11/fast/text/font-selection-font-loading-api-parse-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220285 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd tests for NeverDestroyed
commit-queue@webkit.org [Fri, 4 Aug 2017 19:04:47 +0000 (19:04 +0000)]
Add tests for NeverDestroyed
https://bugs.webkit.org/show_bug.cgi?id=175146

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-08-04
Reviewed by Darin Adler.

* CMakeLists.txt:
* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
New files and sort.

* TestWebKitAPI/Tests/WTF/Logger.h:
* TestWebKitAPI/Tests/WTF/Logger.cpp:
(TestWebKitAPI::log):
(TestWebKitAPI::takeLogStr):
* TestWebKitAPI/Tests/WTF/RefLogger.h:
* TestWebKitAPI/Tests/WTF/RefLogger.cpp:
(TestWebKitAPI::log): Deleted.
(TestWebKitAPI::takeLogStr): Deleted.
Extract log() / takeLogStr() from RefLogger to a general Logger.h/cpp.

* TestWebKitAPI/Tests/WTF/LifecycleLogger.h:
* TestWebKitAPI/Tests/WTF/LifecycleLogger.cpp: Added.
(TestWebKitAPI::LifecycleLogger::LifecycleLogger):
(TestWebKitAPI::LifecycleLogger::operator=):
(TestWebKitAPI::LifecycleLogger::~LifecycleLogger):
(TestWebKitAPI::LifecycleLogger::setName):
(TestWebKitAPI::TEST):
Add a class that logs construction / assignment / modifications.

* TestWebKitAPI/Tests/WTF/NeverDestroyed.cpp: Added.
(TestWebKitAPI::TEST):
(TestWebKitAPI::list):
Test construction behavior is as expected both directly and using makeNeverDestroyed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220284 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agominification logic is not implemented for ios builds
lforschler@apple.com [Fri, 4 Aug 2017 19:03:27 +0000 (19:03 +0000)]
minification logic is not implemented for ios builds
<rdar://problem/33726561>

Reviewed by Dean Johnson

* BuildSlaveSupport/built-product-archive:
(minifyDirectory): refactor minifySource -> minifyDirectory
(archiveBuiltProduct): add ios minification logic
(minifySource): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220283 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[ESNext] Async iteration - update feature.json
gskachkov@gmail.com [Fri, 4 Aug 2017 19:01:00 +0000 (19:01 +0000)]
[ESNext] Async iteration - update feature.json
https://bugs.webkit.org/show_bug.cgi?id=175197

Reviewed by Yusuke Suzuki.

Update feature.json to add status of the Async Iteration

* features.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220282 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: REGRESSION (r220233): Check for null pointer passed to WebGLRenderingC...
mattbaker@apple.com [Fri, 4 Aug 2017 18:57:03 +0000 (18:57 +0000)]
Web Inspector: REGRESSION (r220233): Check for null pointer passed to WebGLRenderingContextBase::deleteProgram
https://bugs.webkit.org/show_bug.cgi?id=175196
<rdar://problem/33727603>

Reviewed by Devin Rousso.

* html/canvas/WebGLRenderingContextBase.cpp:
(WebCore::WebGLRenderingContextBase::deleteProgram):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220281 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFixed loading feature status page queries and anchor link URLs
jond@apple.com [Fri, 4 Aug 2017 18:32:09 +0000 (18:32 +0000)]
Fixed loading feature status page queries and anchor link URLs
https://bugs.webkit.org/show_bug.cgi?id=175156

Reviewed by Simon Fraser.

* wp-content/themes/webkit/status.php:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220280 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r220271.
jlewis3@apple.com [Fri, 4 Aug 2017 18:28:23 +0000 (18:28 +0000)]
Unreviewed, rolling out r220271.

Rolling out due to Layout Test failing on iOS Simulator.

Reverted changeset:

"Remove STREAMS_API compilation guard"
https://bugs.webkit.org/show_bug.cgi?id=175165
http://trac.webkit.org/changeset/220271

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220279 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Cleanup] Remove ScriptGlobalObject
weinig@apple.com [Fri, 4 Aug 2017 17:59:54 +0000 (17:59 +0000)]
[Cleanup] Remove ScriptGlobalObject
https://bugs.webkit.org/show_bug.cgi?id=175173

Reviewed by Darin Adler.

Source/WebCore:

ScriptGlobalObject's two functions were only being used in
four places. Three of those uses (ScriptGlobalObject::set in
InspectorFrontendClientLocal, WebInspectorUI, and RemoteWebInspectorUI)
were merged into the new function addSelfToGlobalObjectInWorld on
InspectorFrontendHost. The remaining function (ScriptGlobalObject::get
in InspectorFrontendHost) was easily inlined.

* CMakeLists.txt:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSBindingsAllInOne.cpp:
* bindings/js/ScriptGlobalObject.cpp: Removed.
* bindings/js/ScriptGlobalObject.h: Removed.
Removed ScriptGlobalObject.

* inspector/InspectorFrontendClientLocal.cpp:
(WebCore::InspectorFrontendClientLocal::windowObjectCleared):
Remove call to ScriptGlobalObject::set and use addSelfToGlobalObjectInWorld instead.

* inspector/InspectorFrontendHost.h:
* inspector/InspectorFrontendHost.cpp:
(WebCore::InspectorFrontendHost::showContextMenu):
Inline ScriptGlobalObject::get.

(WebCore::InspectorFrontendHost::addSelfToGlobalObjectInWorld):
Add helper which inlines ScriptGlobalObject::set and works
as a helper for the three inspector frontends.

Source/WebKit:

* WebProcess/WebPage/RemoteWebInspectorUI.cpp:
(WebKit::RemoteWebInspectorUI::windowObjectCleared):
* WebProcess/WebPage/WebInspectorUI.cpp:
(WebKit::WebInspectorUI::windowObjectCleared):
Remove call to ScriptGlobalObject::set and use addSelfToGlobalObjectInWorld instead.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220278 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoPROGRESSION? Multiple imported/w3c/web-platform-tests/fetch/api/ test have started...
cdumez@apple.com [Fri, 4 Aug 2017 16:45:28 +0000 (16:45 +0000)]
PROGRESSION? Multiple imported/w3c/web-platform-tests/fetch/api/ test have started to fail.
https://bugs.webkit.org/show_bug.cgi?id=175061

Unreviewed, mark Fetch/Cors tests as failing on wk2 ElCapitan only, as those tests seem to be passing
everywhere else.

* platform/mac-wk2/TestExpectations:
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220276 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Cocoa] Web Automation: copy JavaScript atoms to WebKit.framework private headers
bburg@apple.com [Fri, 4 Aug 2017 16:44:44 +0000 (16:44 +0000)]
[Cocoa] Web Automation: copy JavaScript atoms to WebKit.framework private headers
https://bugs.webkit.org/show_bug.cgi?id=175088
<rdar://problem/33685818>

Reviewed by Joseph Pecoraro.

* WebKit.xcodeproj/project.pbxproj:
- Add a Copy Files phase to WebKit.framework.
- Copy atoms to the atoms/ directory in WebKit.framework/PrivateHeaders/.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220275 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r220268.
ryanhaddad@apple.com [Fri, 4 Aug 2017 16:17:57 +0000 (16:17 +0000)]
Unreviewed, rolling out r220268.

This change caused assertion failures on macOS and iOS Debug
WK2.

Reverted changeset:

"Resource Load Statistics: Report user interaction
immediately, but only when needed"
https://bugs.webkit.org/show_bug.cgi?id=175090
http://trac.webkit.org/changeset/220268

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220274 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoNetworkLoad should always invoke its redirect completion handler
akling@apple.com [Fri, 4 Aug 2017 16:02:00 +0000 (16:02 +0000)]
NetworkLoad should always invoke its redirect completion handler
https://bugs.webkit.org/show_bug.cgi?id=175179
<rdar://problem/33464999>

Reviewed by Chris Dumez.

Make sure the redirect completion handler is always invoked,
just like we already did for the response completion handler.

* NetworkProcess/NetworkLoad.cpp:
(WebKit::NetworkLoad::~NetworkLoad):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220273 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMark beacon-basic-string.html as slow.
cdumez@apple.com [Fri, 4 Aug 2017 15:39:29 +0000 (15:39 +0000)]
Mark beacon-basic-string.html as slow.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220272 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove STREAMS_API compilation guard
commit-queue@webkit.org [Fri, 4 Aug 2017 15:20:35 +0000 (15:20 +0000)]
Remove STREAMS_API compilation guard
https://bugs.webkit.org/show_bug.cgi?id=175165

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-04
Reviewed by Darin Adler.

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

No change of behavior.

* Configurations/FeatureDefines.xcconfig:
* Modules/fetch/FetchBody.cpp:
(WebCore::FetchBody::consumeAsStream):
* Modules/fetch/FetchBody.h:
* Modules/fetch/FetchBodyOwner.cpp:
(WebCore::FetchBodyOwner::isDisturbedOrLocked const):
(WebCore::FetchBodyOwner::blobLoadingSucceeded):
(WebCore::FetchBodyOwner::blobLoadingFailed):
(WebCore::FetchBodyOwner::blobChunk):
* Modules/fetch/FetchBodyOwner.h:
* Modules/fetch/FetchResponse.cpp:
(WebCore::FetchResponse::BodyLoader::didSucceed):
(WebCore::FetchResponse::BodyLoader::didFail):
(WebCore::FetchResponse::BodyLoader::didReceiveData):
* Modules/fetch/FetchResponse.h:
* Modules/fetch/FetchResponse.idl:
* Modules/fetch/FetchResponse.js:
(initializeFetchResponse):
* Modules/fetch/FetchResponseSource.cpp:
* Modules/fetch/FetchResponseSource.h:
* Modules/streams/ByteLengthQueuingStrategy.idl:
* Modules/streams/ByteLengthQueuingStrategy.js:
* Modules/streams/CountQueuingStrategy.idl:
* Modules/streams/CountQueuingStrategy.js:
* Modules/streams/ReadableByteStreamController.idl:
* Modules/streams/ReadableByteStreamController.js:
* Modules/streams/ReadableByteStreamInternals.js:
* Modules/streams/ReadableStream.idl:
* Modules/streams/ReadableStream.js:
* Modules/streams/ReadableStreamBYOBReader.idl:
* Modules/streams/ReadableStreamBYOBReader.js:
* Modules/streams/ReadableStreamBYOBRequest.idl:
* Modules/streams/ReadableStreamBYOBRequest.js:
* Modules/streams/ReadableStreamDefaultController.idl:
* Modules/streams/ReadableStreamDefaultController.js:
* Modules/streams/ReadableStreamDefaultReader.idl:
* Modules/streams/ReadableStreamDefaultReader.js:
* Modules/streams/ReadableStreamInternals.js:
* Modules/streams/ReadableStreamSource.h:
* Modules/streams/ReadableStreamSource.idl:
* Modules/streams/StreamInternals.js:
* Modules/streams/WritableStream.idl:
* Modules/streams/WritableStream.js:
* Modules/streams/WritableStreamInternals.js:
* bindings/js/JSDOMGlobalObject.cpp:
(WebCore::isReadableByteStreamAPIEnabled):
(WebCore::JSDOMGlobalObject::addBuiltinGlobals):
* bindings/js/JSReadableStreamPrivateConstructors.cpp:
* bindings/js/JSReadableStreamPrivateConstructors.h:
* bindings/js/JSReadableStreamSourceCustom.cpp:
* bindings/js/ReadableStreamDefaultController.cpp:
* bindings/js/ReadableStreamDefaultController.h:
* page/RuntimeEnabledFeatures.h:
* testing/Internals.cpp:
* testing/Internals.h:
* testing/Internals.idl:

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

* Configurations/FeatureDefines.xcconfig:
* WebProcess/InjectedBundle/InjectedBundle.cpp:
(WebKit::InjectedBundle::overrideBoolPreferenceForTestRunner):

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:
* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):
* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220271 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoEnable ServiceWorkers at runtime for WebKitTestRunner.
beidson@apple.com [Fri, 4 Aug 2017 15:16:34 +0000 (15:16 +0000)]
Enable ServiceWorkers at runtime for WebKitTestRunner.
https://bugs.webkit.org/show_bug.cgi?id=175174

Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

* web-platform-tests/background-fetch/interfaces-expected.txt:
* web-platform-tests/background-fetch/interfaces-worker.https-expected.txt:
* web-platform-tests/fetch/api/policies/referrer-no-referrer-service-worker.https-expected.txt:
* web-platform-tests/fetch/api/policies/referrer-origin-service-worker.https-expected.txt:
* web-platform-tests/fetch/api/policies/referrer-origin-when-cross-origin-service-worker.https-expected.txt:
* web-platform-tests/fetch/api/policies/referrer-unsafe-url-service-worker.https-expected.txt:
* web-platform-tests/html/webappapis/scripting/events/messageevent-constructor.https-expected.txt:
* web-platform-tests/html/webappapis/scripting/processing-model-2/integration-with-the-javascript-agent-formalism/canblock-serviceworker.https-expected.txt:
* web-platform-tests/streams/byte-length-queuing-strategy.serviceworker.https-expected.txt:
* web-platform-tests/streams/count-queuing-strategy.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/close-propagation-backward.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/close-propagation-forward.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/error-propagation-backward.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/error-propagation-forward.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/flow-control.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/general.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/multiple-propagation.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/pipe-through.serviceworker.https-expected.txt:
* web-platform-tests/streams/piping/transform-streams.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-byte-streams/general.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/bad-strategies.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/bad-underlying-sources.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/brand-checks.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/cancel.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/count-queuing-strategy-integration.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/default-reader.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/floating-point-total-queue-size.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/garbage-collection.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/general.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/pipe-through.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/readable-stream-reader.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/tee.serviceworker.https-expected.txt:
* web-platform-tests/streams/readable-streams/templated.serviceworker.https-expected.txt:

Source/WebKit:

* UIProcess/WebPreferences.cpp:
(WebKit::WebPreferences::enableAllExperimentalFeatures):

LayoutTests:

* fast/dom/navigator-detached-no-crash-expected.txt:
* platform/mac-wk1/fast/dom/navigator-detached-no-crash-expected.txt: Copied from LayoutTests/fast/dom/navigator-detached-no-crash-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/background-fetch/interfaces-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/background-fetch/interfaces-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/background-fetch/interfaces-worker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/background-fetch/interfaces-worker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/policies/referrer-no-referrer-service-worker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/fetch/api/policies/referrer-no-referrer-service-worker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/policies/referrer-origin-service-worker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/fetch/api/policies/referrer-origin-service-worker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/policies/referrer-origin-when-cross-origin-service-worker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/fetch/api/policies/referrer-origin-when-cross-origin-service-worker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/policies/referrer-unsafe-url-service-worker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/fetch/api/policies/referrer-unsafe-url-service-worker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/html/webappapis/scripting/events/messageevent-constructor.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/html/webappapis/scripting/events/messageevent-constructor.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/html/webappapis/scripting/processing-model-2/integration-with-the-javascript-agent-formalism/canblock-serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/html/webappapis/scripting/processing-model-2/integration-with-the-javascript-agent-formalism/canblock-serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/byte-length-queuing-strategy.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/byte-length-queuing-strategy.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/count-queuing-strategy.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/count-queuing-strategy.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/close-propagation-backward.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/close-propagation-backward.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/close-propagation-forward.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/close-propagation-forward.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/error-propagation-backward.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/error-propagation-backward.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/error-propagation-forward.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/error-propagation-forward.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/flow-control.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/flow-control.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/general.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/general.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/multiple-propagation.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/multiple-propagation.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/pipe-through.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/pipe-through.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/piping/transform-streams.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/transform-streams.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-byte-streams/general.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/general.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/bad-strategies.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/bad-strategies.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/bad-underlying-sources.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/bad-underlying-sources.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/brand-checks.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/brand-checks.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/cancel.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/cancel.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/count-queuing-strategy-integration.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/count-queuing-strategy-integration.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/default-reader.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/default-reader.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/floating-point-total-queue-size.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/floating-point-total-queue-size.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/garbage-collection.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/garbage-collection.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/general.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/general.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/pipe-through.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/piping/pipe-through.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/readable-stream-reader.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/readable-stream-reader.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/tee.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/tee.serviceworker.https-expected.txt.
* platform/mac-wk1/imported/w3c/web-platform-tests/streams/readable-streams/templated.serviceworker.https-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/streams/readable-streams/templated.serviceworker.https-expected.txt.
* platform/mac/js/dom/global-constructors-attributes-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220270 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EME][GStreamer] Register ClearKey CDMFactory
zandobersek@gmail.com [Fri, 4 Aug 2017 15:04:40 +0000 (15:04 +0000)]
[EME][GStreamer] Register ClearKey CDMFactory
https://bugs.webkit.org/show_bug.cgi?id=175136

Reviewed by Xabier Rodriguez-Calvar.

Source/WebCore:

Register the ClearKey CDM factory in CDMFactoryGStreamer. A CDMFactoryClearKey
singleton object is introduced and used for that registration.

The basic CDMFactoryClearKey and CDMPrivateClearKey methods are implemented,
marking `org.w3.clearkey` as the supported key system and `keyids` as the
supported init data type. Additional logic around key system configurations,
distinctive identifiers, persistent state and related restrictions and
requirements is implemented.

This improves the ClearKey EME tests a bit, now progressing to the point of
failing with a NotAllowedError exception due to the CDMInstance object failing
to properly initialize because of missing implementation.

No new tests -- relevant tests have underlying baselines updated to reflect
changes in behavior.

* platform/encryptedmedia/clearkey/CDMClearKey.cpp:
(WebCore::CDMFactoryClearKey::singleton):
(WebCore::CDMFactoryClearKey::supportsKeySystem):
(WebCore::CDMPrivateClearKey::supportsInitDataType const):
(WebCore::CDMPrivateClearKey::supportsConfiguration const):
(WebCore::CDMPrivateClearKey::supportsConfigurationWithRestrictions const):
(WebCore::CDMPrivateClearKey::supportsSessionTypeWithConfiguration const):
(WebCore::CDMPrivateClearKey::supportsRobustness const):
(WebCore::CDMPrivateClearKey::distinctiveIdentifiersRequirement const):
(WebCore::CDMPrivateClearKey::persistentStateRequirement const):
* platform/encryptedmedia/clearkey/CDMClearKey.h:
* platform/graphics/gstreamer/eme/CDMFactoryGStreamer.cpp:
(WebCore::CDMFactory::platformRegisterFactories):

LayoutTests:

Update WPE baselines for EME ClearKey tests following some advancements in
ClearKey support.

* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-events-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-events-session-closed-event-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-generate-request-disallowed-input-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-invalid-license-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-keystatuses-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-keystatuses-multiple-sessions-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-clear-encrypted-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-encrypted-clear-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-encrypted-clear-sources-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-events-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-sequential-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multikey-sequential-readyState-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-multisession-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-after-src-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-after-update-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-immediately-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-setMediaKeys-onencrypted-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-two-videos-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-playback-temporary-waitingforkey-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-requestmediakeysystemaccess-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-reset-src-after-setmediakeys-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-again-after-playback-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-again-after-resetting-src-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-at-same-time-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-multiple-times-with-the-same-mediakeys-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-setmediakeys-to-multiple-video-elements-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-syntax-mediakeys-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-syntax-mediakeysession-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-syntax-mediakeysystemaccess-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-update-disallowed-input-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-mp4-waiting-for-a-key-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-not-callable-after-createsession-expected.txt:
* platform/wpe/imported/w3c/web-platform-tests/encrypted-media/clearkey-update-non-ascii-input-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220269 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoResource Load Statistics: Report user interaction immediately, but only when needed
wilander@apple.com [Fri, 4 Aug 2017 14:20:02 +0000 (14:20 +0000)]
Resource Load Statistics: Report user interaction immediately, but only when needed
https://bugs.webkit.org/show_bug.cgi?id=175090
<rdar://problem/33685546>

Reviewed by Chris Dumez.

Source/WebCore:

Test: http/tests/loading/resourceLoadStatistics/user-interaction-only-reported-once-within-short-period-of-time.html

* loader/ResourceLoadObserver.cpp:
(WebCore::ResourceLoadObserver::ResourceLoadObserver):
(WebCore::ResourceLoadObserver::logUserInteractionWithReducedTimeResolution):
    Now tells the UI process immediately but also records that it has
    done so to avoid doing it when not needed.
(WebCore::ResourceLoadObserver::scheduleNotificationIfNeeded):
    Conditional throttling gone, now always throttles.
(WebCore::ResourceLoadObserver::notifyObserver):
    Renamed from ResourceLoadObserver::notificationTimerFired().
(WebCore::ResourceLoadObserver::clearState):
    New function to allow the test runner to reset the web process'
    statistics state now that we keep track of whether or not we've
    reported user interaction to the UI process.
(WebCore::ResourceLoadObserver::setShouldThrottleObserverNotifications): Deleted.
(WebCore::ResourceLoadObserver::notificationTimerFired): Deleted.
* loader/ResourceLoadObserver.h:
(): Deleted.
* testing/Internals.cpp:
(WebCore::Internals::resetToConsistentState):
(WebCore::Internals::setResourceLoadStatisticsShouldThrottleObserverNotifications): Deleted.
    No longer needed since user interaction is always communicated
    immediately.
* testing/Internals.h:
* testing/Internals.idl:

Source/WebKit:

* WebProcess/InjectedBundle/API/c/WKBundle.cpp:
(WKBundleClearResourceLoadStatistics):
    Test infrastructure. Ends up calling
    WebCore::ResourceLoadObserver::clearState().
* WebProcess/InjectedBundle/API/c/WKBundlePrivate.h:

Tools:

* WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:
(WTR::InjectedBundle::beginTesting):
    Now calls WebCore::ResourceLoadObserver::clearState().

LayoutTests:

* http/tests/loading/resourceLoadStatistics/user-interaction-in-cross-origin-sub-frame.html:
    Now no longer needs to disable throttling since reports of
    user interaction happen immediately (when needed).
* http/tests/loading/resourceLoadStatistics/user-interaction-only-reported-once-within-short-period-of-time-expected.txt: Added.
* http/tests/loading/resourceLoadStatistics/user-interaction-only-reported-once-within-short-period-of-time.html: Added.
* platform/mac-wk2/TestExpectations:
    user-interaction-only-reported-once-within-short-period-of-time.html marked as [ Pass ].

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220268 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoNetwork cache should be usable as non-singleton
antti@apple.com [Fri, 4 Aug 2017 13:25:02 +0000 (13:25 +0000)]
Network cache should be usable as non-singleton
https://bugs.webkit.org/show_bug.cgi?id=175139

Reviewed by Sam Weinig.

We might want to use it as a non-singleton in the future (for example as a backend for the cache API).

This patch makes NetworkCache::Cache and NetworkCache::Storage refcounted objects and takes
care to ref them properly during asynchronous operations.

The patch doesn't actually create any non-shared instances, it just adds the capability.

* NetworkProcess/NetworkConnectionToWebProcess.cpp:
(WebKit::NetworkConnectionToWebProcess::storeDerivedDataToCache):
* NetworkProcess/NetworkProcess.cpp:
(WebKit::fetchDiskCacheEntries):
(WebKit::clearDiskCacheEntries):
(WebKit::NetworkProcess::setCacheModel):
* NetworkProcess/NetworkProcess.h:
(WebKit::NetworkProcess::cache):

    Move the shared cache instance to the network process singleton.

* NetworkProcess/NetworkResourceLoader.cpp:
(WebKit::NetworkResourceLoader::NetworkResourceLoader):

    Include shared cache as a reffed member for non-ephemeral instances.

(WebKit::NetworkResourceLoader::canUseCache const):
(WebKit::NetworkResourceLoader::retrieveCacheEntry):
(WebKit::NetworkResourceLoader::abort):
(WebKit::NetworkResourceLoader::didReceiveResponse):
(WebKit::NetworkResourceLoader::willSendRedirectedRequest):
(WebKit::NetworkResourceLoader::tryStoreAsCacheEntry):
(WebKit::NetworkResourceLoader::didRetrieveCacheEntry):
* NetworkProcess/NetworkResourceLoader.h:
* NetworkProcess/cache/NetworkCache.cpp:
(WebKit::NetworkCache::Cache::open):

    Open now returns null if it fails.
    Add RegisterNotify option to set up notify trigger for the shared instance.

(WebKit::NetworkCache::Cache::Cache):
(WebKit::NetworkCache::Cache::~Cache):

    Add destructor.

(WebKit::NetworkCache::dumpFileChanged):
(WebKit::NetworkCache::Cache::setCapacity):
(WebKit::NetworkCache::Cache::retrieve):

    Protect the cache instance during asynchronous operations.

(WebKit::NetworkCache::Cache::store):
(WebKit::NetworkCache::Cache::storeRedirect):
(WebKit::NetworkCache::Cache::remove):
(WebKit::NetworkCache::Cache::traverse):
(WebKit::NetworkCache::Cache::dumpContentsToFile):
(WebKit::NetworkCache::Cache::clear):
(WebKit::NetworkCache::Cache::recordsPath const):
(WebKit::NetworkCache::Cache::retrieveData):
(WebKit::NetworkCache::Cache::storeData):
(WebKit::NetworkCache::singleton): Deleted.
(WebKit::NetworkCache::Cache::initialize): Deleted.
* NetworkProcess/cache/NetworkCache.h:
(WebKit::NetworkCache::Cache::canUseSharedMemoryForBodyData const):
(WebKit::NetworkCache::Cache::isEnabled const): Deleted.

    Remove isEnabled() state as a cache object now always represents an enabled cache.

* NetworkProcess/cache/NetworkCacheEntry.cpp:
(WebKit::NetworkCache::Entry::initializeShareableResourceHandleFromStorageRecord const):
* NetworkProcess/cache/NetworkCacheSpeculativeLoad.cpp:
(WebKit::NetworkCache::SpeculativeLoad::willSendRedirectedRequest):
(WebKit::NetworkCache::SpeculativeLoad::didReceiveResponse):
(WebKit::NetworkCache::SpeculativeLoad::didFinishLoading):
* NetworkProcess/cache/NetworkCacheStatistics.cpp:
(WebKit::NetworkCache::Statistics::initialize):
(WebKit::NetworkCache::Statistics::shrinkIfNeeded):
* NetworkProcess/cache/NetworkCacheStorage.cpp:
(WebKit::NetworkCache::Storage::ReadOperation::ReadOperation):
(WebKit::NetworkCache::Storage::WriteOperation::WriteOperation):
(WebKit::NetworkCache::Storage::TraverseOperation::TraverseOperation):

    Operations now ref the storage. They are already deleted in the main thread so
    proper destruction is taken care of.

(WebKit::NetworkCache::Storage::open):
(WebKit::NetworkCache::Storage::~Storage):
(WebKit::NetworkCache::Storage::synchronize):

    This and other asynchronous methods now protect the Storage instance.

(WebKit::NetworkCache::Storage::remove):
(WebKit::NetworkCache::Storage::retrieve):
(WebKit::NetworkCache::Storage::store):
(WebKit::NetworkCache::Storage::traverse):
(WebKit::NetworkCache::Storage::clear):
(WebKit::NetworkCache::Storage::shrink):
(WebKit::NetworkCache::Storage::deleteOldVersions):
* NetworkProcess/cache/NetworkCacheStorage.h:
* NetworkProcess/cocoa/NetworkProcessCocoa.mm:
(WebKit::NetworkProcess::platformInitializeNetworkProcessCocoa):
(WebKit::NetworkProcess::clearDiskCache):
* NetworkProcess/cocoa/NetworkSessionCocoa.mm:
(WebKit::NetworkSessionCocoa::NetworkSessionCocoa):
* NetworkProcess/soup/NetworkProcessSoup.cpp:
(WebKit::NetworkProcess::platformInitializeNetworkProcess):
(WebKit::NetworkProcess::clearDiskCache):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220267 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Test gardening around MOUSE_CURSOR_SCALE.
clopez@igalia.com [Fri, 4 Aug 2017 10:12:53 +0000 (10:12 +0000)]
[GTK] Test gardening around MOUSE_CURSOR_SCALE.
https://bugs.webkit.org/show_bug.cgi?id=109469

Patch by Ms2ger <Ms2ger@igalia.com> on 2017-08-04
Reviewed by Carlos Alberto Lopez Perez.

Remove obsolete expectation for cursor-parsing-image-set.html; it may have
started passing in r209396.

Mark mouse-cursor-image-set.html as failing rather than using an incorrect
-expected file.

* platform/gtk/TestExpectations:
* platform/gtk/fast/events/mouse-cursor-image-set-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220266 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoJSC test wasm/js-api/test_memory_constructor.js should be skipped on memoryLimited
clopez@igalia.com [Fri, 4 Aug 2017 10:01:15 +0000 (10:01 +0000)]
JSC test wasm/js-api/test_memory_constructor.js should be skipped on memoryLimited
https://bugs.webkit.org/show_bug.cgi?id=175150

Unreviewed test gardening.

* wasm/js-api/test_memory_constructor.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220265 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EME] Push CDMFactory into the platform layer
zandobersek@gmail.com [Fri, 4 Aug 2017 09:54:42 +0000 (09:54 +0000)]
[EME] Push CDMFactory into the platform layer
https://bugs.webkit.org/show_bug.cgi?id=175129

Reviewed by Xabier Rodriguez-Calvar.

This is a follow-up to r219678 that moved the majority of CDM abstraction
classes into the platform layer, but missed the CDMFactory class.

The CDMFactory abstraction is now also placed in the platform layer. Only
change to the interface is that the createCDM() method can't accept a CDM
object reference anymore since that class is cemented into the WebCore
layer, and no current implementation used it anyway.

Additionally, the static Vector object of registered factories is moved
under the CDMFactory class, along with the register and unregister
functions. The platformRegisterFactories() function is added to allow for
platform-specific factory registrations to occur when the registered
factories are queried for the first time. Empty implementation for this
function is provided for non-GStreamer platforms, while for GStreamer
the implementation is kept in CDMFactoryGStreamer.cpp. It's still empty
for now, but it will register the ClearKey factory there in the near
future.

No new tests -- none affected, only refactoring.

* CMakeLists.txt:
* Modules/encryptedmedia/CDM.cpp:
(WebCore::createCDMPrivateForKeySystem):
(WebCore::CDM::supportsKeySystem):
(WebCore::CDM::CDM):
(): Deleted.
(WebCore::CDM::registerCDMFactory): Deleted.
(WebCore::CDM::unregisterCDMFactory): Deleted.
* Modules/encryptedmedia/CDM.h:
(WebCore::CDMFactory::~CDMFactory): Deleted.
* PlatformWPE.cmake:
* platform/GStreamer.cmake:
* platform/encryptedmedia/CDMFactory.cpp: Added.
(WebCore::CDMFactory::registerFactory):
(WebCore::CDMFactory::unregisterFactory):
(WebCore::CDMFactory::platformRegisterFactories):
* platform/encryptedmedia/CDMFactory.h: Added.
(WebCore::CDMFactory::~CDMFactory):
* platform/encryptedmedia/clearkey/CDMClearKey.cpp:
(WebCore::CDMFactoryClearKey::createCDM):
* platform/encryptedmedia/clearkey/CDMClearKey.h:
* platform/encryptedmedia/gstreamer/CDMFactoryGStreamer.cpp: Added.
(WebCore::CDMFactory::platformRegisterFactories):
* testing/MockCDMFactory.cpp:
(WebCore::m_weakPtrFactory):
(WebCore::MockCDMFactory::unregister):
(WebCore::MockCDMFactory::createCDM):
* testing/MockCDMFactory.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220264 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed GTK+ gardening. Update test expectations and layout test baselines
zandobersek@gmail.com [Fri, 4 Aug 2017 09:08:27 +0000 (09:08 +0000)]
Unreviewed GTK+ gardening. Update test expectations and layout test baselines
for Web Crypto tests now that the implementation is complete.

* platform/gtk/TestExpectations:
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/encrypt_decrypt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/encrypt_decrypt/aes_cbc.worker-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/encrypt_decrypt/test_aes_cbc.https-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/generateKey/successes.worker-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/import_export: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/import_export/rsa_importKey.worker-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/import_export/test_rsa_importKey.https-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/rsa_pkcs.worker-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/rsa_pss.worker-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/test_rsa_pkcs.https-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/test_rsa_pss.https-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/wrapKey_unwrapKey: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/wrapKey_unwrapKey/test_wrapKey_unwrapKey.https-expected.txt: Added.
* platform/gtk/imported/w3c/web-platform-tests/WebCryptoAPI/wrapKey_unwrapKey/wrapKey_unwrapKey.worker-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220263 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EsNext] Async iteration - Add feature flag
gskachkov@gmail.com [Fri, 4 Aug 2017 08:51:12 +0000 (08:51 +0000)]
[EsNext] Async iteration - Add feature flag
https://bugs.webkit.org/show_bug.cgi?id=166694

Reviewed by Yusuke Suzuki.

Add feature flag to JSC to switch on/off Async Iterator

* runtime/Options.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220262 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoScrollingTreeOverflowScrollingNodeIOS uses the wrong fixed position rectangle
fred.wang@free.fr [Fri, 4 Aug 2017 08:08:17 +0000 (08:08 +0000)]
ScrollingTreeOverflowScrollingNodeIOS uses the wrong fixed position rectangle
https://bugs.webkit.org/show_bug.cgi?id=175135

Patch by Frederic Wang <fwang@igalia.com> on 2017-08-04
Reviewed by Simon Fraser.

Source/WebCore:

This patch modifies ScrollingTreeOverflowScrollingNodeIOS::updateChildNodesAfterScroll so
that it uses the fixed position rectangle relative of the first frame ancestor instead of
the one of the main frame. This makes it consistent with ScrollingTreeFrameScrollingNodeIOS
and RenderLayerCompositor. This fixes some flickering issues on iOS.

Test: fast/scrolling/ios/fixed-inside-overflow-inside-iframe.html

* page/scrolling/ScrollingTreeFrameScrollingNode.h:
(WebCore::ScrollingTreeFrameScrollingNode::fixedPositionRect): Helper function to get the
fixed position rect to use for that frame.
* page/scrolling/ScrollingTreeNode.cpp:
(WebCore::ScrollingTreeNode::enclosingFrameNode const): Helper function to get the enclosing
frame for this scrolling node or null if there is none.
* page/scrolling/ScrollingTreeNode.h: Declare enclosingFrameNode.

Source/WebKit:

This patch modifies ScrollingTreeOverflowScrollingNodeIOS::updateChildNodesAfterScroll so
that it uses the fixed position rectangle relative of the first frame ancestor instead of
the one of the main frame. This makes it consistent with ScrollingTreeFrameScrollingNodeIOS
and RenderLayerCompositor. This fixes some flickering issues on iOS.

* UIProcess/Scrolling/ios/ScrollingTreeOverflowScrollingNodeIOS.mm:
(WebKit::ScrollingTreeOverflowScrollingNodeIOS::updateChildNodesAfterScroll): Use the fixed
position rect of a non-main frame if there is one.

LayoutTests:

This patch adds a new test for a position:fixed element inside an overflow node inside an
iframe. When scrolling the overflow node, the position of such an element should remain fixed
relative to the inner frame. Before that change, ScrollingTreeOverflowScrollingNodeIOS used
to take the main frame as a reference instead, causing the element to flicker and even to
disappear when the user scrolls that overflow node. We add a reftest to verify that the
element is visible and positioned at the correct location when the user scrolls.

* fast/scrolling/ios/fixed-inside-overflow-inside-iframe-expected.html: Added.
* fast/scrolling/ios/fixed-inside-overflow-inside-iframe.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220261 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed. Removing redundant NotImplemented.h header inclusions
zandobersek@gmail.com [Fri, 4 Aug 2017 07:56:17 +0000 (07:56 +0000)]
Unreviewed. Removing redundant NotImplemented.h header inclusions
and cleaning up whitespace issues in libgcrypt-specific CryptoKeyEC
and CryptoKeyRSA implementation files.

* crypto/gcrypt/CryptoKeyECGCrypt.cpp:
* crypto/gcrypt/CryptoKeyRSAGCrypt.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220260 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed WPE gardening. Update test expectations and layout test baselines
zandobersek@gmail.com [Fri, 4 Aug 2017 07:52:43 +0000 (07:52 +0000)]
Unreviewed WPE gardening. Update test expectations and layout test baselines
for Web Crypto tests now that the implementation is complete.

* platform/wpe/TestExpectations:
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/generateKey/successes.worker-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/import_export: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/import_export/rsa_importKey.worker-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/import_export/test_rsa_importKey.https-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/rsa_pkcs.worker-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/rsa_pss.worker-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/test_rsa_pkcs.https-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/sign_verify/test_rsa_pss.https-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/wrapKey_unwrapKey: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/wrapKey_unwrapKey/test_wrapKey_unwrapKey.https-expected.txt: Added.
* platform/wpe/imported/w3c/web-platform-tests/WebCryptoAPI/wrapKey_unwrapKey/wrapKey_unwrapKey.worker-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220259 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove unnecesary call to status bar SPI.
commit-queue@webkit.org [Fri, 4 Aug 2017 07:36:50 +0000 (07:36 +0000)]
Remove unnecesary call to status bar SPI.
https://bugs.webkit.org/show_bug.cgi?id=175176
rdar://problem/20887306

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-04
Reviewed by Darin Adler.

No new tests because no behavior change.

This removes an obsolete call to SPI.

* platform/ios/VideoFullscreenInterfaceAVKit.mm:
(VideoFullscreenInterfaceAVKit::cleanupFullscreen):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220258 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GCrypt] Implement CryptoKeyEC PKCS#8 imports
zandobersek@gmail.com [Fri, 4 Aug 2017 06:13:23 +0000 (06:13 +0000)]
[GCrypt] Implement CryptoKeyEC PKCS#8 imports
https://bugs.webkit.org/show_bug.cgi?id=173647

Reviewed by Jiewen Tan.

Source/WebCore:

No new tests -- affected tests are now passing and are unskipped.

Implement libgcrypt-based support for PKCS#8 imports of EC keys.

Existing libtasn1 utilities are used to achieve this. First, the provided key data
is decoded against the PrivateKeyInfo ASN.1 definition. First, the version member
of that structure is validated, followed by the algorithm member. The latter is
also properly tested depending on this being an import of an ECDSA or ECDH key.

Data of the parameters member is decoded against the ECParameters ASN.1 definition,
and the namedCurve object identifier is validated, making sure it represents a
valid EC curve and that this curve maches the one specified for the import
operation.

Data of the privateKey member is decoded against the ECPrivateKey ASN.1 definition.
The version member of that structure is properly validated. The optional parameters
member of that structure is already decoded against the ECParameters ASN.1
definition. If present, it is checked to contain a valid EC curve identifier that
matches the specified curve.

The optional publicKey member of the ECPrivateKey structure is validated, testing
that its data matches in size an uncompressed EC point, and that the first byte
of this data is 0x04, as expected for an uncompressed EC point.

What's left is the private key data on the initial ECPrivateKey structure. That
data is retrieved and validated, making sure its size matches the size of the
specified curve. The `private-key` s-expression is then constructed, embedding
the curve name and the validated private key data. This s-expression is then used
to construct an EC context.

If the optional publicKey data was provided, it's used to set the `q` parameter
for this EC context. Otherwise, the value for `q` is computed on-the-fly for the
specified EC and the provided private key. The `q` point is then tested through
the gcry_mpi_ec_curve_point() function, making sure that the derived point is
indeed located on the given EC.

Finally, with the private key properly validated, a new CryptoKeyEC object is
constructed, using the `private-key` s-expression and the parameters that were
specified for this import operation.

* crypto/gcrypt/CryptoKeyECGCrypt.cpp:
(WebCore::CryptoKeyEC::platformImportPkcs8):
* crypto/gcrypt/GCryptUtilities.h:

LayoutTests:

* platform/wpe/TestExpectations:
Unskip passing Web Crypto tests that cover PKCS#8 imports of EC keys.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220253 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix parsing of <meta http-equiv=refresh> to allow time starting with a '.' without...
cdumez@apple.com [Fri, 4 Aug 2017 04:37:10 +0000 (04:37 +0000)]
Fix parsing of <meta http-equiv=refresh> to allow time starting with a '.' without a leading 0
https://bugs.webkit.org/show_bug.cgi?id=175132

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Re-sync WPT tests from upstream c8bf1bbe9296. This extends test coverage.

* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/allow-scripts-flag-changing-1-expected.txt:
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/allow-scripts-flag-changing-2-expected.txt:
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/moving-documents-expected.txt: Removed.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/parsing-expected.txt:
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/parsing.html:
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/remove-from-document-expected.txt: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/remove-from-document.html: Renamed from LayoutTests/imported/w3c/web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/moving-documents.html.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/refresh.sub.html: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/w3c-import.log:
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/w3c-import.log:

Source/WebCore:

Fix parsing of <meta http-equiv=refresh> to allow time starting with a '.', without
a leading 0. This is as per https://github.com/whatwg/html/pull/2852.

The latest spec is at:
- https://html.spec.whatwg.org/multipage/semantics.html#attr-meta-http-equiv-refresh

Test: imported/w3c/web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/parsing.html

* html/parser/HTMLParserIdioms.cpp:
(WebCore::parseHTTPRefreshInternal):

LayoutTests:

* TestExpectations:
Mark as flaky tests that used to not run because they were missing a subresource. Now that I imported
this subresource, the tests run but are failing. When they fail, those tests are flaky due to the lines
they log.

* tests-options.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220252 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoJSContext Inspector: Scripts sometimes do not show in resources tab
commit-queue@webkit.org [Fri, 4 Aug 2017 03:57:29 +0000 (03:57 +0000)]
JSContext Inspector: Scripts sometimes do not show in resources tab
https://bugs.webkit.org/show_bug.cgi?id=175153
<rdar://problem/33708683>

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-08-03
Reviewed by Matt Baker.

* UserInterface/Views/ResourceSidebarPanel.js:
(WI.ResourceSidebarPanel.prototype.initialLayout):
When the ResourceSidebar is lazily created, be sure to add any scripts
to the sidebar that are not backed by Resources.

(WI.ResourceSidebarPanel.prototype._scriptWasAdded):
(WI.ResourceSidebarPanel.prototype._addScript):
Extract so it can be used outside of an event handler.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220251 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove ENABLE(WEB_SOCKET) guards
bburg@apple.com [Fri, 4 Aug 2017 02:41:16 +0000 (02:41 +0000)]
Remove ENABLE(WEB_SOCKET) guards
https://bugs.webkit.org/show_bug.cgi?id=167044

Reviewed by Joseph Pecoraro.

.:

* Source/cmake/OptionsMac.cmake:
* Source/cmake/OptionsWin.cmake:
* Source/cmake/WebKitFeatures.cmake:
* Source/cmake/tools/vsprops/FeatureDefines.props:
* Source/cmake/tools/vsprops/FeatureDefinesCairo.props:

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

* Configurations/FeatureDefines.xcconfig:
* Modules/websockets/ThreadableWebSocketChannel.cpp:
* Modules/websockets/ThreadableWebSocketChannel.h:
* Modules/websockets/ThreadableWebSocketChannelClientWrapper.cpp:
* Modules/websockets/ThreadableWebSocketChannelClientWrapper.h:
* Modules/websockets/WebSocket.cpp:
* Modules/websockets/WebSocket.h:
* Modules/websockets/WebSocket.idl:
* Modules/websockets/WebSocketChannel.cpp:
* Modules/websockets/WebSocketChannel.h:
* Modules/websockets/WebSocketChannelClient.h:
* Modules/websockets/WebSocketDeflateFramer.cpp:
* Modules/websockets/WebSocketDeflateFramer.h:
* Modules/websockets/WebSocketDeflater.cpp:
* Modules/websockets/WebSocketDeflater.h:
* Modules/websockets/WebSocketExtensionDispatcher.cpp:
* Modules/websockets/WebSocketExtensionDispatcher.h:
* Modules/websockets/WebSocketExtensionParser.cpp:
* Modules/websockets/WebSocketExtensionParser.h:
* Modules/websockets/WebSocketExtensionProcessor.h:
* Modules/websockets/WebSocketFrame.cpp:
* Modules/websockets/WebSocketFrame.h:
* Modules/websockets/WebSocketHandshake.cpp:
* Modules/websockets/WebSocketHandshake.h:
* Modules/websockets/WorkerThreadableWebSocketChannel.cpp:
* Modules/websockets/WorkerThreadableWebSocketChannel.h:
* dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::idbConnectionProxy):
(WebCore::Document::socketProvider):
* dom/Document.h:
* dom/ScriptExecutionContext.h:
* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::didSendWebSocketFrameImpl):
* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::didSendWebSocketFrame):
* inspector/InspectorNetworkAgent.cpp:
* inspector/InspectorNetworkAgent.h:
* page/RuntimeEnabledFeatures.cpp:
(WebCore::RuntimeEnabledFeatures::webSocketEnabled const):
* page/RuntimeEnabledFeatures.h:
* page/SocketProvider.cpp:
* page/SocketProvider.h:
* workers/WorkerGlobalScope.cpp:
(WebCore::WorkerGlobalScope::WorkerGlobalScope):
* workers/WorkerGlobalScope.h:
* workers/WorkerMessagingProxy.cpp:
(WebCore::WorkerMessagingProxy::startWorkerGlobalScope):
* workers/WorkerThread.cpp:
(WebCore::WorkerThread::WorkerThread):
(WebCore::WorkerThread::socketProvider):
* workers/WorkerThread.h:

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

* Configurations/FeatureDefines.xcconfig:
* WebProcess/Network/WebSocketProvider.cpp:
* WebProcess/Network/WebSocketProvider.h:

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:
* WebView/WebPreferencesPrivate.h:

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* Scripts/webkitperl/FeatureList.pm:
* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220250 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove LayoutUnit dependency in TextStream
don.olmstead@sony.com [Fri, 4 Aug 2017 02:37:53 +0000 (02:37 +0000)]
Remove LayoutUnit dependency in TextStream
https://bugs.webkit.org/show_bug.cgi?id=175110

Reviewed by Zalan Bujtas.

No new tests. No change in behavior.

* CMakeLists.txt:
* WebCore.xcodeproj/project.pbxproj:
* platform/LayoutUnit.cpp: Added.
(WebCore::operator<<):
* platform/LayoutUnit.h:
* platform/text/TextStream.cpp:
* platform/text/TextStream.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220249 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoImprove WebKitLegacy video fullscreen animation begin and end rects.
commit-queue@webkit.org [Fri, 4 Aug 2017 01:58:33 +0000 (01:58 +0000)]
Improve WebKitLegacy video fullscreen animation begin and end rects.
https://bugs.webkit.org/show_bug.cgi?id=175152
rdar://problem/32840576

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-03
Reviewed by Eric Carlson.

No new tests, becuase this change has no effect on the DOM.

This change uses different rects for fullscreen animation to prevent the animation
from failing, and to improve the aesthetics of the animation.

* platform/mac/WebVideoFullscreenController.mm:
(frameExpandedToRatioOfFrame):
(-[WebVideoFullscreenController enterFullscreen:]):
(-[WebVideoFullscreenController exitFullscreen]):
(-[WebVideoFullscreenWindow animateFromRect:toRect:withSubAnimation:controllerAction:]):
(constrainFrameToRatioOfFrame): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220248 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EME][Mac] SecureStop left on disk in Private Browsing mode.
jer.noble@apple.com [Fri, 4 Aug 2017 01:13:03 +0000 (01:13 +0000)]
[EME][Mac] SecureStop left on disk in Private Browsing mode.
https://bugs.webkit.org/show_bug.cgi?id=175162

Reviewed by Eric Carlson.

Return an empty string from mediaKeysStorageDirectory() when the page indicates that storage should
be ephemeral(). Previously, an empty string in this case would be treated as an error. Instead, treat
an empty string as valid, and do not try to store or retrieve session information to disk in that case.

* Modules/encryptedmedia/legacy/WebKitMediaKeySession.cpp:
(WebCore::WebKitMediaKeySession::mediaKeysStorageDirectory const):
* platform/graphics/avfoundation/objc/CDMSessionAVContentKeySession.mm:
(WebCore::CDMSessionAVContentKeySession::releaseKeys):
(WebCore::CDMSessionAVContentKeySession::update):
(WebCore::CDMSessionAVContentKeySession::generateKeyReleaseMessage):
(WebCore::CDMSessionAVContentKeySession::contentKeySession):
* platform/graphics/avfoundation/objc/CDMSessionMediaSourceAVFObjC.mm:
(WebCore::CDMSessionMediaSourceAVFObjC::storagePath const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220247 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r219850): run-benchmark script broken on Linux
clopez@igalia.com [Fri, 4 Aug 2017 01:12:52 +0000 (01:12 +0000)]
REGRESSION(r219850): run-benchmark script broken on Linux
https://bugs.webkit.org/show_bug.cgi?id=175126

Reviewed by Stephanie Lewis.

The run-benchmark script dynamically generates the list of supported
browsers and platforms (currently Linux and OSX) by loading all
python files from Tools/Scripts/webkitpy/benchmark_runner/browser_driver
and getting the browser_name and platform variables from the
classes defined there.

This means that this classes should not raise an exception when
loaded on other platforms or otherwise they will broke the whole
script. Its fine if they raise an exception when executing any of
the methods they implement, but not when just loading/importing
the class.

Move the argument variable definitions that call on the platform
specific OSXBrowserDriver._screen_size() function from beeing
variables that are evaluated when loading the file, to be functions
that are only evaluated when the actual functionality needs to be
executed.

* Scripts/webkitpy/benchmark_runner/browser_driver/osx_chrome_driver.py:
(OSXChromeDriver.launch_url):
(OSXChromeCanaryDriver.launch_url):
(create_args):
(create_chrome_options):
(create_window_size_arg):
* Scripts/webkitpy/benchmark_runner/browser_driver/osx_firefox_driver.py:
(OSXFirefoxDriver.launch_url):
(OSXFirefoxNightlyDriver.launch_url):
(OSXFirefoxNightlyDriver.launch_driver):
(create_args):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220246 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMultiple Layout tests from web-platform-tests/beacon/ are timing out.
cdumez@apple.com [Fri, 4 Aug 2017 01:07:24 +0000 (01:07 +0000)]
Multiple Layout tests from web-platform-tests/beacon/ are timing out.
https://bugs.webkit.org/show_bug.cgi?id=175076
<rdar://problem/33704752>

Reviewed by Alexey Proskuryakov.

LayoutTests/imported/w3c:

Rebaseline tests that are now passing.

* web-platform-tests/fetch/api/cors/cors-basic.any-expected.txt:
* web-platform-tests/fetch/api/cors/cors-basic.any.worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-no-preflight.any-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin.any-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin.any.worker-expected.txt:

Source/WebKit:

* NetworkProcess/NetworkLoad.cpp:
(WebKit::NetworkLoad::completeAuthenticationChallenge):
In the NETWORK_SESSION code path, we did not ask the client about server trust evaluation
when the clientCredentialPolicy was CannotAskClientForCredentials. This is because the
same delegate is used for HTTP authentication and server trust evaluation in the
NETWORK_SESSION code path. To align both code paths, we now ask the client about server
trust evaluation even if the policy CannotAskClientForCredentials. This allows WKTR
to trust certificates for localhost / 127.0.0.1 unconditionally and consistently.

LayoutTests:

* platform/ios-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt: Removed.
* platform/ios-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt: Removed.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic.any-expected.txt: Removed.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic.any.worker-expected.txt: Removed.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/cors/cors-no-preflight.any-expected.txt: Removed.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin.any-expected.txt: Removed.
* platform/mac-wk1/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin.any.worker-expected.txt: Removed.
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt: Removed.
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt: Removed.
Drop platform-specific expectations as those tests are now passing everywhere.

* platform/wk2/TestExpectations:
Unskip tests that are now passing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220245 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Fetch API] Add support for Request keepalive getter
commit-queue@webkit.org [Fri, 4 Aug 2017 01:00:13 +0000 (01:00 +0000)]
[Fetch API] Add support for Request keepalive getter
https://bugs.webkit.org/show_bug.cgi?id=175151

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-03
Reviewed by Chris Dumez.

LayoutTests/imported/w3c:

* web-platform-tests/fetch/api/request/request-idl-expected.txt:
* web-platform-tests/fetch/api/request/request-idl.html:
* web-platform-tests/fetch/api/request/request-keepalive-expected.txt: Added.
* web-platform-tests/fetch/api/request/request-keepalive.html: Added.

Source/WebCore:

Test: imported/w3c/web-platform-tests/fetch/api/request/request-keepalive.html

Adding keepalive as a fetch option.
Adding initialization and getter of keepalive into FetchRequest.

* Modules/fetch/FetchRequest.cpp:
(WebCore::buildOptions):
* Modules/fetch/FetchRequest.h:
* Modules/fetch/FetchRequest.idl:
* loader/FetchOptions.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220244 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[PAL] Move spi/cf directory into PAL
commit-queue@webkit.org [Fri, 4 Aug 2017 00:20:37 +0000 (00:20 +0000)]
[PAL] Move spi/cf directory into PAL
https://bugs.webkit.org/show_bug.cgi?id=175057

Patch by Yoshiaki Jitsukawa <jitsu@rd.scei.sony.co.jp> on 2017-08-03
Reviewed by Antti Koivisto.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj:
* loader/cocoa/DiskCacheMonitorCocoa.mm:
* loader/cocoa/SubresourceLoaderCocoa.mm:
* loader/mac/ResourceLoaderMac.mm:
* platform/cf/CoreMediaSoftLink.cpp:
* platform/cf/CoreMediaSoftLink.h:
* platform/mac/PluginBlacklist.mm:
* platform/mac/WebCoreNSStringExtras.mm:
* platform/mac/WebGLBlacklist.mm:
* platform/mediastream/mac/CoreAudioCaptureSource.cpp:
* platform/network/NetworkStorageSession.h:
* platform/network/cf/CookieJarCFNet.cpp:
* platform/network/cf/CredentialStorageCFNet.cpp:
* platform/network/cf/ResourceHandleCFNet.cpp:
* platform/network/cf/ResourceHandleCFURLConnectionDelegate.cpp:
* platform/network/cf/ResourceHandleCFURLConnectionDelegateWithOperationQueue.cpp:
* platform/network/cf/ResourceRequestCFNet.cpp:
* platform/network/cf/ResourceRequestCFNet.h:
* platform/network/cf/ResourceResponse.h:
* platform/network/cf/ResourceResponseCFNet.cpp:
* platform/network/cf/SocketStreamHandleImplCFNet.cpp:
* platform/network/cf/SynchronousResourceHandleCFURLConnectionDelegate.cpp:
* platform/network/cocoa/CookieStorageObserver.h:
* platform/network/cocoa/CredentialCocoa.h:
* platform/network/cocoa/NetworkStorageSessionCocoa.mm:
* platform/network/cocoa/ResourceRequestCocoa.mm:
* platform/network/cocoa/ResourceResponseCocoa.mm:
* platform/network/ios/ResourceRequestIOS.mm:
* platform/network/mac/CookieJarMac.mm:
* platform/network/mac/FormDataStreamMac.mm:
* platform/network/mac/ResourceHandleMac.mm:
* platform/network/mac/ResourceRequestMac.mm:
* platform/network/mac/WebCoreResourceHandleAsDelegate.mm:
* platform/network/mac/WebCoreResourceHandleAsOperationQueueDelegate.mm:
* platform/network/mac/WebCoreURLResponse.h:
* testing/cocoa/WebArchiveDumpSupport.mm:

Source/WebCore/PAL:

* PAL.xcodeproj/project.pbxproj:
* pal/spi/cf/CFLocaleSPI.h: Renamed from Source/WebCore/platform/spi/cf/CFLocaleSPI.h.
* pal/spi/cf/CFNetworkConnectionCacheSPI.h: Renamed from Source/WebCore/platform/spi/cf/CFNetworkConnectionCacheSPI.h.
* pal/spi/cf/CFNetworkSPI.h: Renamed from Source/WebCore/platform/spi/cf/CFNetworkSPI.h.
* pal/spi/cf/CFUtilitiesSPI.h: Renamed from Source/WebCore/platform/spi/cf/CFUtilitiesSPI.h.
* pal/spi/cf/CoreAudioSPI.h: Renamed from Source/WebCore/platform/spi/cf/CoreAudioSPI.h.
* pal/spi/cf/CoreMediaSPI.h: Renamed from Source/WebCore/platform/spi/cf/CoreMediaSPI.h.

Source/WebKit:

* NetworkProcess/cocoa/NetworkDataTaskCocoa.mm:
* NetworkProcess/cocoa/NetworkProcessCocoa.mm:
* NetworkProcess/cocoa/NetworkSessionCocoa.mm:
* NetworkProcess/ios/NetworkProcessIOS.mm:
* NetworkProcess/mac/NetworkLoadMac.mm:
* NetworkProcess/mac/NetworkProcessMac.mm:
* Shared/cf/CookieStorageUtilsCF.h:
* Shared/mac/ChildProcessMac.mm:
* Shared/mac/CookieStorageShim.mm:
* UIProcess/API/Cocoa/WKHTTPCookieStore.mm:
* UIProcess/API/Cocoa/WKProcessPool.mm:
* UIProcess/Cocoa/WebProcessPoolCocoa.mm:
* UIProcess/Plugins/mac/PluginProcessProxyMac.mm:
* UIProcess/WebsiteData/Cocoa/WebsiteDataStoreCocoa.mm:
* UIProcess/WebsiteData/WebsiteDataRecord.cpp:
* UIProcess/WebsiteData/WebsiteDataStore.h:
* UIProcess/mac/WebCookieManagerProxyMac.mm:
* WebProcess/WebCoreSupport/mac/WebFrameNetworkingContext.mm:
* WebProcess/cocoa/WebProcessCocoa.mm:

Source/WebKitLegacy/mac:

* Plugins/Hosted/HostedNetscapePluginStream.mm:
* Plugins/WebNetscapePluginStream.mm:
* WebCoreSupport/WebFrameNetworkingContext.mm:
* WebView/WebPreferences.mm:
* WebView/WebView.mm:

Tools:

* Scripts/webkitpy/style/checker.py:
* Scripts/webkitpy/style/checker_unittest.py:
(GlobalVariablesTest.test_path_rules_specifier):

Ignore "readability/naming/underscores" style errors for sources
under the WebCore/PAL/pal/spi directory.

* TestWebKitAPI/Tests/WebKit2Cocoa/CookieAcceptPolicy.mm:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220243 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove FETCH_API compilation guard
commit-queue@webkit.org [Thu, 3 Aug 2017 23:54:36 +0000 (23:54 +0000)]
Remove FETCH_API compilation guard
https://bugs.webkit.org/show_bug.cgi?id=175154

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-03
Reviewed by Chris Dumez.

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

Source/WebCore:

No change of behavior.

* Configurations/FeatureDefines.xcconfig:
* Modules/fetch/DOMWindowFetch.cpp:
* Modules/fetch/DOMWindowFetch.h:
* Modules/fetch/DOMWindowFetch.idl:
* Modules/fetch/FetchBody.cpp:
* Modules/fetch/FetchBody.h:
* Modules/fetch/FetchBody.idl:
* Modules/fetch/FetchBodyConsumer.cpp:
* Modules/fetch/FetchBodyConsumer.h:
* Modules/fetch/FetchBodyOwner.cpp:
* Modules/fetch/FetchBodyOwner.h:
* Modules/fetch/FetchHeaders.cpp:
* Modules/fetch/FetchHeaders.h:
* Modules/fetch/FetchHeaders.idl:
* Modules/fetch/FetchInternals.js:
* Modules/fetch/FetchLoader.cpp:
* Modules/fetch/FetchLoader.h:
* Modules/fetch/FetchLoaderClient.h:
* Modules/fetch/FetchRequest.cpp:
* Modules/fetch/FetchRequest.h:
* Modules/fetch/FetchRequest.idl:
* Modules/fetch/FetchResponse.cpp:
* Modules/fetch/FetchResponse.h:
* Modules/fetch/FetchResponse.idl:
* Modules/fetch/FetchResponse.js:
* Modules/fetch/FetchResponseSource.cpp:
* Modules/fetch/FetchResponseSource.h:
* Modules/fetch/WorkerGlobalScopeFetch.cpp:
* Modules/fetch/WorkerGlobalScopeFetch.h:
* Modules/fetch/WorkerGlobalScopeFetch.idl:
* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::fetchAPIEnabled const):

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

* Configurations/FeatureDefines.xcconfig:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:
* WebView/WebPreferences.mm:
(+[WebPreferences initialize]):
* WebView/WebView.mm:
(-[WebView _preferencesChanged:]):

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220241 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agohttps://bugs.webkit.org/show_bug.cgi?id=174718
lforschler@apple.com [Thu, 3 Aug 2017 23:17:41 +0000 (23:17 +0000)]
https://bugs.webkit.org/show_bug.cgi?id=174718

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220238 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: add button to open Inspector^2
drousso@apple.com [Thu, 3 Aug 2017 23:10:57 +0000 (23:10 +0000)]
Web Inspector: add button to open Inspector^2
https://bugs.webkit.org/show_bug.cgi?id=175108

Reviewed by Brian Burg.

Source/WebCore:

This patch just exposes a function to the inspector page. No new functionality was added.

* inspector/InspectorFrontendHost.idl:
* inspector/InspectorFrontendHost.h:
* inspector/InspectorFrontendHost.cpp:
(WebCore::InspectorFrontendHost::inspectInspector):

Source/WebInspectorUI:

* UserInterface/Debug/Bootstrap.js:
(updateDebugUI):
(WI.runBootstrapOperations):

* UserInterface/Views/ButtonToolbarItem.js:
(WI.ButtonToolbarItem):
(WI.ButtonToolbarItem.prototype.get label): Deleted.
(WI.ButtonToolbarItem.prototype.set label): Deleted.
* UserInterface/Views/ButtonToolbarItem.css:
(.toolbar .item.button):
(.toolbar .item.button:not(.disabled):active):
(.toolbar .item.button:not(.disabled):matches(:focus, .activate.activated)):
(.toolbar .item.button:not(.disabled):active:matches(:focus, .activate.activated)):
(.toolbar .item.button > .glyph):
(.toolbar .item.button:not(.disabled):active > .glyph): Deleted.
(.toolbar .item.button:not(.disabled):matches(:focus, .activate.activated) > .glyph): Deleted.
(.toolbar .item.button:not(.disabled):active:matches(:focus, .activate.activated) > .glyph): Deleted.
(.toolbar .item.button > .label): Deleted.
* UserInterface/Views/ActivateButtonToolbarItem.js:
(WI.ActivateButtonToolbarItem):
(WI.ActivateButtonToolbarItem.prototype.get label): Deleted.
(WI.ActivateButtonToolbarItem.prototype.set label): Deleted.
* UserInterface/Base/Main.js:
(WI.contentLoaded):
Remove unused `label` parameter from Toolbar objects.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220235 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Instrument WebGLProgram created/deleted
mattbaker@apple.com [Thu, 3 Aug 2017 22:31:44 +0000 (22:31 +0000)]
Web Inspector: Instrument WebGLProgram created/deleted
https://bugs.webkit.org/show_bug.cgi?id=175059

Reviewed by Devin Rousso.

Source/JavaScriptCore:

Extend the Canvas protocol with types/events for tracking WebGLPrograms.

* inspector/protocol/Canvas.json:

Source/WebCore:

Tests: inspector/canvas/shaderProgram-add-remove-webgl.html
       inspector/canvas/shaderProgram-add-remove-webgl2.html

This patch adds instrumentation to WebGLRenderingContextBase for tracking
WebGLPrograms. A new helper class, InspectorShaderProgram, is used by
the CanvasAgent to hold related data.

* CMakeLists.txt:
* WebCore.xcodeproj/project.pbxproj:

* html/canvas/WebGLRenderingContextBase.cpp:
(WebCore::WebGLRenderingContextBase::createProgram):
(WebCore::WebGLRenderingContextBase::deleteProgram):

* inspector/InspectorCanvasAgent.cpp:
(WebCore::InspectorCanvasAgent::enable):
(WebCore::InspectorCanvasAgent::frameNavigated):
(WebCore::InspectorCanvasAgent::didCreateProgram):
(WebCore::InspectorCanvasAgent::willDeleteProgram):
(WebCore::InspectorCanvasAgent::clearCanvasData):
(WebCore::InspectorCanvasAgent::unbindCanvas):
(WebCore::InspectorCanvasAgent::unbindProgram):
(WebCore::InspectorCanvasAgent::assertInspectorProgram):
(WebCore::InspectorCanvasAgent::findInspectorProgram):
* inspector/InspectorCanvasAgent.h:

* inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::didCreateCSSCanvasImpl):
(WebCore::InspectorInstrumentation::didChangeCSSCanvasClientNodesImpl):
(WebCore::InspectorInstrumentation::didCreateCanvasRenderingContextImpl):
(WebCore::InspectorInstrumentation::didChangeCanvasMemoryImpl):
(WebCore::InspectorInstrumentation::recordCanvasActionImpl):
(WebCore::InspectorInstrumentation::didFinishRecordingCanvasFrameImpl):
(WebCore::InspectorInstrumentation::didCreateProgramImpl):
(WebCore::InspectorInstrumentation::willDeleteProgramImpl):

* inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::recordCanvasActionImpl):
(WebCore::InspectorInstrumentation::didCreateCSSCanvas):
(WebCore::InspectorInstrumentation::didChangeCSSCanvasClientNodes):
(WebCore::InspectorInstrumentation::didCreateCanvasRenderingContext):
(WebCore::InspectorInstrumentation::didChangeCanvasMemory):
(WebCore::InspectorInstrumentation::recordCanvasAction):
(WebCore::InspectorInstrumentation::didFinishRecordingCanvasFrame):
(WebCore::InspectorInstrumentation::didCreateProgram):
(WebCore::InspectorInstrumentation::willDeleteProgram):

* inspector/InspectorShaderProgram.cpp: Added.
(WebCore::InspectorShaderProgram::create):
(WebCore::InspectorShaderProgram::InspectorShaderProgram):
(WebCore::InspectorShaderProgram::context const):
* inspector/InspectorShaderProgram.h: Added.

Source/WebInspectorUI:

This patch adds frontend support for shader program instrumentation.
The frontend creates a ShaderProgram model object for each WebGLProgram.
Since only canvases with a WebGL context have programs, the Canvas model
object does not contain any logic specific to programs. CanvasManager
dispatches program added/removed events, and the parent Canvas can be
accessed from ShaderProgram but not the other way around.

* UserInterface/Controllers/CanvasManager.js:
(WI.CanvasManager):
(WI.CanvasManager.prototype.get shaderPrograms):
(WI.CanvasManager.prototype.canvasRemoved):
(WI.CanvasManager.prototype.programCreated):
(WI.CanvasManager.prototype.programDeleted):
(WI.CanvasManager.prototype._mainResourceDidChange):
(WI.CanvasManager.prototype._dispatchShaderProgramRemoved):

* UserInterface/Main.html:

* UserInterface/Models/Canvas.js:
(WI.Canvas.prototype.nextShaderProgramDisplayNumber):
(WI.Canvas):

* UserInterface/Models/ShaderProgram.js: Added.
(WI.ShaderProgram):
(WI.ShaderProgram.prototype.get identifier):
(WI.ShaderProgram.prototype.get canvas):
(WI.ShaderProgram.prototype.get displayName):

* UserInterface/Protocol/CanvasObserver.js:
(WI.CanvasObserver.prototype.programCreated):
(WI.CanvasObserver.prototype.programDeleted):
(WI.CanvasObserver):

* UserInterface/Test.html:

LayoutTests:

Add tests for CanvasManager shader program events and ShaderProgram model object.
WebGL and WebGL2 contexts are tested separately based on platform support.

* inspector/canvas/resources/shaderProgram-utilities.js: Added.
(createProgram):
(deleteProgram):
(deleteContext):
(TestPage.registerInitializer.awaitProgramAdded):
(TestPage.registerInitializer):
(TestPage.registerInitializer.window.initializeTestSuite):
(TestPage.registerInitializer.window.addSimpleTestCase):
(TestPage.registerInitializer.window.addParentCanvasRemovedTestCase):

* inspector/canvas/shaderProgram-add-remove-webgl-expected.txt: Added.
* inspector/canvas/shaderProgram-add-remove-webgl.html: Added.
* inspector/canvas/shaderProgram-add-remove-webgl2-expected.txt: Added.
* inspector/canvas/shaderProgram-add-remove-webgl2.html: Added.

* platform/gtk/TestExpectations:
* platform/mac/TestExpectations:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220233 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoJSContext Inspector: Recording tab should not be available in New Tab picker
commit-queue@webkit.org [Thu, 3 Aug 2017 22:18:11 +0000 (22:18 +0000)]
JSContext Inspector: Recording tab should not be available in New Tab picker
https://bugs.webkit.org/show_bug.cgi?id=175155

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-08-03
Reviewed by Brian Burg.

* UserInterface/Views/RecordingTabContentView.js:
(WI.RecordingTabContentView.isTabAllowed):
Only allow the Recording Tab if we have a CanvasAgent.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220232 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, rolling out r220209.
jlewis3@apple.com [Thu, 3 Aug 2017 20:27:49 +0000 (20:27 +0000)]
Unreviewed, rolling out r220209.

This caused internal build failures.

Reverted changeset:

"Use MPAVRoutingController instead of deprecated versions."
https://bugs.webkit.org/show_bug.cgi?id=175063
http://trac.webkit.org/changeset/220209

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220226 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoImport WPT service worker tests
commit-queue@webkit.org [Thu, 3 Aug 2017 20:21:29 +0000 (20:21 +0000)]
Import WPT service worker tests
https://bugs.webkit.org/show_bug.cgi?id=175053

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-03
Reviewed by Brady Eidson.

LayoutTests/imported/w3c:

Importing service-worker tests up to cfdfb48329b20e19b6492a317ac5181a99506fd2.

* resources/resource-files.json:
* resources/import-expectations.json:
* web-platform-tests/service-workers/:

LayoutTests:

* TestExpectations: skipping service worker tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220223 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Automation: consider file extensions in the "accept" attribute when deciding...
bburg@apple.com [Thu, 3 Aug 2017 20:19:43 +0000 (20:19 +0000)]
Web Automation: consider file extensions in the "accept" attribute when deciding if a file can be uploaded
https://bugs.webkit.org/show_bug.cgi?id=175081

Reviewed by Joseph Pecoraro.

* UIProcess/Automation/WebAutomationSession.cpp:
(WebKit::fileCanBeAcceptedForUpload):
(WebKit::WebAutomationSession::handleRunOpenPanel):
In cases where a file has an extension, try to match it against the
accepted extensions list. Give up if there's no extension. Otherwise,
proceed as normal to infer a MIME type based on the extension.
In cases where extension-less files are inside hidden folders, the MIME
inference will fail as well.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220222 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK][WKE] Pass the --memory-limited option on the GTK and WPE buildbots for the...
clopez@igalia.com [Thu, 3 Aug 2017 19:37:13 +0000 (19:37 +0000)]
[GTK][WKE] Pass the --memory-limited option on the GTK and WPE buildbots for the JSC tests.
https://bugs.webkit.org/show_bug.cgi?id=175140

Reviewed by Alexey Proskuryakov.

We are having lately issues with JSC tests causing problems on
the GTK+ and WPE bots due to the high amount of memory some tests
need to run.

The best thing we can do now is to workaround this by disabling
all the tests marked as memoryLimited on the GTK and WPE bots.
We may revise this on the future.

* BuildSlaveSupport/build.webkit.org-config/master.cfg:
(RunJavaScriptCoreTests.start):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220221 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd SW IDLs and stub out basic functionality.
beidson@apple.com [Thu, 3 Aug 2017 19:21:43 +0000 (19:21 +0000)]
Add SW IDLs and stub out basic functionality.
https://bugs.webkit.org/show_bug.cgi?id=175115

Reviewed by Chris Dumez.

.:

* Source/cmake/WebKitFeatures.cmake:
* Source/cmake/tools/vsprops/FeatureDefines.props:
* Source/cmake/tools/vsprops/FeatureDefinesCairo.props:

Source/JavaScriptCore:

* Configurations/FeatureDefines.xcconfig:

* runtime/CommonIdentifiers.h:

Source/WebCore:

No new tests (Currently no behavior change).

Overall note: This feature is EnabledAtRuntime as opposed to EnabledBySetting because
the Settings-based code generation is completely broken for non-Document contexts,
whereas the RuntimeEnabledFeatures-based generation is not.

* CMakeLists.txt:
* Configurations/FeatureDefines.xcconfig:
* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:

* bindings/scripts/preprocess-idls.pl: Handle the new global scope c'tor file.

* bindings/js/JSServiceWorkerContainerCustom.cpp: Added.
(WebCore::JSServiceWorkerContainer::ready const):

* bindings/js/JSWorkerGlobalScopeBase.cpp:
(WebCore::toJSWorkerGlobalScope): Refactor to handle both types of derived workers.
(WebCore::toJSServiceWorkerGlobalScope):
* bindings/js/JSWorkerGlobalScopeBase.h:

* dom/EventNames.h:
* dom/EventTargetFactory.in:

* features.json: Change status of feature.

* page/Navigator.idl:
* page/NavigatorBase.cpp:
(WebCore::NavigatorBase::serviceWorker):
* page/NavigatorBase.h:
* page/NavigatorServiceWorker.idl: Added.

* page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::serviceWorkerEnabled const):
(WebCore::RuntimeEnabledFeatures::setServiceWorkerEnabled):

* workers/ServiceWorker.cpp: Added.
(WebCore::ServiceWorker::postMessage):
(WebCore::ServiceWorker::~ServiceWorker):
(WebCore::ServiceWorker::scriptURL const):
(WebCore::ServiceWorker::state const):
(WebCore::ServiceWorker::eventTargetInterface const):
(WebCore::ServiceWorker::scriptExecutionContext const):
* workers/ServiceWorker.h: Added.
* workers/ServiceWorker.idl: Added.

* workers/ServiceWorkerContainer.cpp: Added.
(WebCore::ServiceWorkerContainer::~ServiceWorkerContainer):
(WebCore::ServiceWorkerContainer::controller const):
(WebCore::ServiceWorkerContainer::ready):
(WebCore::ServiceWorkerContainer::addRegistration):
(WebCore::ServiceWorkerContainer::getRegistration):
(WebCore::ServiceWorkerContainer::getRegistrations):
(WebCore::ServiceWorkerContainer::startMessages):
(WebCore::ServiceWorkerContainer::eventTargetInterface const):
(WebCore::ServiceWorkerContainer::scriptExecutionContext const):
* workers/ServiceWorkerContainer.h: Added.
* workers/ServiceWorkerContainer.idl: Added.

* workers/ServiceWorkerGlobalScope.cpp: Added.
(WebCore::ServiceWorkerGlobalScope::registration):
(WebCore::ServiceWorkerGlobalScope::skipWaiting):
* workers/ServiceWorkerGlobalScope.h: Added.
* workers/ServiceWorkerGlobalScope.idl: Added.

* workers/ServiceWorkerRegistration.cpp: Added.
(WebCore::ServiceWorkerRegistration::~ServiceWorkerRegistration):
(WebCore::ServiceWorkerRegistration::installing):
(WebCore::ServiceWorkerRegistration::waiting):
(WebCore::ServiceWorkerRegistration::active):
(WebCore::ServiceWorkerRegistration::scope const):
(WebCore::ServiceWorkerRegistration::update):
(WebCore::ServiceWorkerRegistration::unregister):
(WebCore::ServiceWorkerRegistration::eventTargetInterface const):
(WebCore::ServiceWorkerRegistration::scriptExecutionContext const):
* workers/ServiceWorkerRegistration.h: Added.
* workers/ServiceWorkerRegistration.idl: Added.

Source/WebCore/PAL:

* Configurations/FeatureDefines.xcconfig:

Source/WebKit:

* Configurations/FeatureDefines.xcconfig:

* Shared/WebPreferencesDefinitions.h:

* UIProcess/WebPreferences.cpp:
(WebKit::WebPreferences::enableAllExperimentalFeatures): Explicitly skip SW for now.
  The ramifications to layouttests are complicated, and we'd like to follow up in a
  separate patch.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

Source/WebKitLegacy/mac:

* Configurations/FeatureDefines.xcconfig:

Source/WTF:

* wtf/FeatureDefines.h:

Tools:

* TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

* Scripts/webkitpy/bindings/main.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220220 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRename ScratchBuffer::activeLengthPtr to addressOfActiveLength.
mark.lam@apple.com [Thu, 3 Aug 2017 19:17:13 +0000 (19:17 +0000)]
Rename ScratchBuffer::activeLengthPtr to addressOfActiveLength.
https://bugs.webkit.org/show_bug.cgi?id=175142
<rdar://problem/33704528>

Reviewed by Filip Pizlo.

The convention in the rest of of JSC for such methods which return the address of
a field is to name them "addressOf<field name>".  We'll rename
ScratchBuffer::activeLengthPtr to be consistent with this convention.

* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileNewArrayWithSpread):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGThunks.cpp:
(JSC::DFG::osrExitGenerationThunkGenerator):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNewArray):
(JSC::FTL::DFG::LowerDFGToB3::compileNewArrayWithSpread):
* ftl/FTLThunks.cpp:
(JSC::FTL::genericGenerationThunkGenerator):
* jit/AssemblyHelpers.cpp:
(JSC::AssemblyHelpers::debugCall):
* jit/ScratchRegisterAllocator.cpp:
(JSC::ScratchRegisterAllocator::preserveUsedRegistersToScratchBufferForCall):
(JSC::ScratchRegisterAllocator::restoreUsedRegistersFromScratchBufferForCall):
* runtime/VM.h:
(JSC::ScratchBuffer::addressOfActiveLength):
(JSC::ScratchBuffer::activeLengthPtr): Deleted.
* wasm/WasmBinding.cpp:
(JSC::Wasm::wasmToJs):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220219 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTest gardening.
commit-queue@webkit.org [Thu, 3 Aug 2017 19:13:43 +0000 (19:13 +0000)]
Test gardening.
https://bugs.webkit.org/show_bug.cgi?id=175137

Unreviewed test gardening.

Patch by Ms2ger <Ms2ger@igalia.com> on 2017-08-03

* platform/gtk/TestExpectations:
* platform/gtk/fast/images/async-image-multiple-clients-repaint-expected.txt: Added.
* platform/gtk/fast/text/atsui-pointtooffset-calls-cg-expected.txt:
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220218 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, build fix for Windows port
utatane.tea@gmail.com [Thu, 3 Aug 2017 18:59:44 +0000 (18:59 +0000)]
Unreviewed, build fix for Windows port
https://bugs.webkit.org/show_bug.cgi?id=175013

* wtf/Threading.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220217 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMerge ThreadHolder to WTF::Thread itself
utatane.tea@gmail.com [Thu, 3 Aug 2017 18:29:16 +0000 (18:29 +0000)]
Merge ThreadHolder to WTF::Thread itself
https://bugs.webkit.org/show_bug.cgi?id=175013

Reviewed by Mark Lam.

Currently, we store ThreadHolder* to the TLS, and ThreadHolder* holds Ref<Thread>.
When we get Thread& from the current thread TLS, we need to dereference the ThreadHolder*.
However, ideally, we can store Thread* directly to the current thread TLS.
While the ThreadHolder design is beautiful, it's worth optimizing by storing Thread* directly
since Thread::current() is so frequently executed.

This patch merges ThreadHolder to Thread. And we now store Thread* directly in the TLS.
When storing it to TLS, we call leakRef() to keep Thread ref count incremented by the TLS.
And when destroying the TLS, we call `deref()` to ensure that Thread* is dereferenced from
the TLS.

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
* wtf/ThreadHolder.cpp: Removed.
* wtf/ThreadHolder.h: Removed.
* wtf/ThreadHolderPthreads.cpp: Removed.
* wtf/ThreadHolderWin.cpp: Removed.
* wtf/Threading.cpp:
(WTF::Thread::entryPoint):
(WTF::initializeThreading):
* wtf/Threading.h:
(WTF::Thread::currentMayBeNull):
(WTF::Thread::current):
* wtf/ThreadingPthreads.cpp:
(WTF::Thread::waitForCompletion):
(WTF::Thread::initializeCurrentTLS):
(WTF::Thread::initializeTLSKey):
(WTF::Thread::initializeTLS):
(WTF::Thread::destructTLS):
(WTF::Thread::createCurrentThread): Deleted.
* wtf/ThreadingWin.cpp:
(WTF::Thread::initializeCurrentTLS):
(WTF::threadMapMutex):
(WTF::Thread::initializeTLSKey):
(WTF::Thread::currentDying):
(WTF::Thread::get):
(WTF::Thread::initializeTLS):
(WTF::Thread::destructTLS):
(WTF::waitForThreadCompletion):
(WTF::Thread::createCurrentThread): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220214 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[WebKit] Sort Xcode project file
commit-queue@webkit.org [Thu, 3 Aug 2017 18:03:47 +0000 (18:03 +0000)]
[WebKit] Sort Xcode project file
https://bugs.webkit.org/show_bug.cgi?id=175122

Patch by Yoshiaki Jitsukawa <jitsu@rd.scei.sony.co.jp> on 2017-08-03
Reviewed by Antti Koivisto.

* WebKit.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220212 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[WebCore] Sort Xcode project files
commit-queue@webkit.org [Thu, 3 Aug 2017 18:03:16 +0000 (18:03 +0000)]
[WebCore] Sort Xcode project files
https://bugs.webkit.org/show_bug.cgi?id=175121

Patch by Yoshiaki Jitsukawa <jitsu@rd.scei.sony.co.jp> on 2017-08-03
Reviewed by Antti Koivisto.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj:

Source/WebCore/PAL:

* PAL.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220211 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoSource/WebCore:
weinig@apple.com [Thu, 3 Aug 2017 17:51:22 +0000 (17:51 +0000)]
Source/WebCore:
[WebIDL] Convert MutationCallback to be a normal generated callback
https://bugs.webkit.org/show_bug.cgi?id=174140

Reviewed by Darin Adler.

To make this work more nicely, I:
- Added the ability to for non-nullable interfaces in sequences to be passed
  via a Ref<> rather than a RefPtr<> as a parameter to a callback function.
  (e.g. callback MyCallback = void (sequence<Foo> foos) will now have the
  signature, CallbackResult<void> handleEvent(const Vector<Ref<Foo>>&) rather
  than CallbackResult<void> handleEvent(const Vector<RefPtr<Foo>>&).
- Added a new extended attribute for callback functions called [CallbackThisObject=Type]
  which allows you to specify that the callback needs a this object in addition
  to its arguments. When specified, the first argument of the C++ implementation
  function will now correspond to the this object, with the remaining arguments
  shifted over one.
- Converted callback objects to all inherit directly from ActiveDOMCallback rather
  than having the generated JS callback derived class inherit from it. This allows
  us to have access to a callback's canInvokeCallback() function anywhere (needed
  for MutationCallback) as well as giving a place to put an optional virtual
  visitJSFunction to allow marking weak callbacks (while not an ideal layering,
  this matches what we do in EventListener). This change requires each callback to
  have a bit more code to import the ActiveDOMCallback's constructor and requires
  non-JS derived callbacks to pass a ScriptExecutionContext (e.g. the Document).

* CMakeLists.txt:
* DerivedSources.make:
* WebCore.xcodeproj/project.pbxproj:
* bindings/js/JSMutationCallback.cpp: Removed.
* bindings/js/JSMutationCallback.h: Removed.
Remove custom JSMutationCallback.h/cpp

* Modules/geolocation/PositionCallback.h:
* Modules/geolocation/PositionErrorCallback.h:
* Modules/notifications/NotificationPermissionCallback.h:
* Modules/webaudio/AudioBufferCallback.h:
* Modules/webdatabase/DatabaseCallback.h:
* Modules/webdatabase/SQLStatementCallback.h:
* Modules/webdatabase/SQLStatementErrorCallback.h:
* Modules/webdatabase/SQLTransactionCallback.h:
* Modules/webdatabase/SQLTransactionErrorCallback.h:
* css/MediaQueryListListener.h:
* dom/NodeFilter.h:
* dom/RequestAnimationFrameCallback.h:
* dom/StringCallback.h:
* fileapi/BlobCallback.h:
* html/VoidCallback.h:
* page/IntersectionObserverCallback.h:
* page/PerformanceObserverCallback.h:
Add ActiveDOMCallback as a base class. Import the ActiveDOMCallback constructor.

* Modules/mediastream/MediaDevicesRequest.cpp:
(WebCore::MediaDevicesRequest::filterDeviceList):
(WebCore::MediaDevicesRequest::start):
* Modules/mediastream/MediaDevicesRequest.h:
Change filterDeviceList to take a Vector of Refs.

* bindings/IDLTypes.h:
Add InnerParameterType and NullableInnerParameterType type hooks
and specialize wrappers to use Ref for InnerParameterType, and RefPtr
for NullableInnerParameterType.

* bindings/js/JSCallbackData.cpp:
* bindings/js/JSCallbackData.h:
Add support for passing a this object and give JSCallbackDataWeak a visitJSFunction
to allow marking the underlying function.

* bindings/js/JSMutationObserverCustom.cpp:
(WebCore::JSMutationObserver::visitAdditionalChildren):
(WebCore::constructJSMutationObserver): Deleted.
Remove the custom constructor and replace it with a custom visitAdditionalChildren
that calls the new ActiveDOMObject's visitJSFunction.

* bindings/scripts/CodeGenerator.pm:
(ParseType):
Add helper to parse a type and cache the result.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateCallbackHeaderContent):
(GenerateCallbackImplementationContent):
(GetJSCallbackDataType): Deleted.
- Add support for [CallbackThisObject]. When [CallbackThisObject] is not specified, use jsUndefined()
  as the this object as specified by WebIDL.
- Stop inheriting from ActiveDOMCallback now that callbacks need to do this themselves.
- Add a visitJSFunction override for weak callback functions which calls into the callback data.

* bindings/scripts/IDLAttributes.json:
Add [CallbackThisObject].

* bindings/scripts/IDLParser.pm:
(ParseType):
Add entry point to parse a single type.

* css/FontFaceSet.h:
Use Ref rather than RefPtr for the faces sequence.

* dom/ActiveDOMCallback.h:
(WebCore::ActiveDOMCallback::visitJSFunction):
Add an optional visitJSFunction virtual function so that derived classes
have a way of marking underlying function objects.

* dom/MutationCallback.h:
Convert to support generation (return a CallbackResult, inherit from ActiveDOMObject).

* dom/MutationCallback.idl: Added.
Added to generate the callback. Uses the new [CallbackThisObject].

* dom/MutationObserver.cpp:
(WebCore::MutationObserver::deliver):
Switch to call idiomatic handleEvent, and pass *this as the first parameter
which will be translated into the this object.

* dom/MutationObserver.h:
(WebCore::MutationObserver::callback):
Expose the callback so it can marked during GC.

* dom/MutationObserver.idl:
Remove CustomConstructor and replace it with a custom mark function.

* dom/NativeNodeFilter.cpp:
* dom/NativeNodeFilter.h:
* inspector/InspectorDatabaseAgent.cpp:
Pass now needed ScriptExecutionContext to non-js based callbacks.

* bindings/scripts/test/JS/JSTestCallbackFunction.cpp:
* bindings/scripts/test/JS/JSTestCallbackFunctionRethrow.cpp:
* bindings/scripts/test/JS/JSTestCallbackFunctionWithThisObject.cpp: Added.
* bindings/scripts/test/JS/JSTestCallbackFunctionWithThisObject.h: Added.
* bindings/scripts/test/JS/JSTestCallbackFunctionWithTypedefs.cpp:
* bindings/scripts/test/JS/JSTestCallbackInterface.cpp:
* bindings/scripts/test/JS/JSTestCallbackInterface.h:
* bindings/scripts/test/JS/JSTestVoidCallbackFunction.cpp:
* bindings/scripts/test/TestCallbackFunctionWithThisObject.idl: Added.
* bindings/scripts/test/TestCallbackInterface.idl:
Add/update tests.

Source/WebKit:
[WebIDL] Convert MutationCallback to be a normal generated callback
https://bugs.webkit.org/show_bug.cgi?id=174140

Reviewed by Darin Adler.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDocument.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNodeFilter.cpp:
* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNodeFilterPrivate.h:
Pass, now necessary, Document to NativeNodeFilter constructor.

Source/WebKitLegacy/mac:
[WebIDL] Convert MutationCallback to be a normal generated callback
https://bugs.webkit.org/show_bug.cgi?id=174140

Reviewed by Darin Adler.

* DOM/DOMDocument.mm:
(-[DOMDocument createNodeIterator:whatToShow:filter:expandEntityReferences:]):
(-[DOMDocument createTreeWalker:whatToShow:filter:expandEntityReferences:]):
Pass, now necessary, Document to NativeNodeFilter constructor.

LayoutTests:
[WebIDL] Convert MutationCallback to be a normal generate callback
https://bugs.webkit.org/show_bug.cgi?id=174140

Reviewed by Darin Adler.

* fast/dom/MutationObserver/mutation-observer-constructor-expected.txt:
Update results for standard error messages.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220210 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUse MPAVRoutingController instead of deprecated versions.
commit-queue@webkit.org [Thu, 3 Aug 2017 17:37:42 +0000 (17:37 +0000)]
Use MPAVRoutingController instead of deprecated versions.
https://bugs.webkit.org/show_bug.cgi?id=175063
Source/WebCore:

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-03
Reviewed by Tim Horton.

No new tests because no behavior change. This uses a different platform class to present
an interface.

Remove deprecated MPAudioVideoRoutingPopoverController and MPAVRoutingSheet
Add MPMediaControlsViewController.

* platform/spi/ios/MediaPlayerSPI.h:

Source/WebKit:

rdar://problem/33301230

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-03
Reviewed by Tim Horton.

Remove dependence on deprecated classes MPAVRoutingSheet and MPAudioVideoRoutingPopoverController.

* UIProcess/ios/forms/WKAirPlayRoutePicker.h:
* UIProcess/ios/forms/WKAirPlayRoutePicker.mm:
(-[WKAirPlayRoutePicker dealloc]):
(-[WKAirPlayRoutePicker show:fromRect:]):
(-[WKAirPlayRoutePicker popoverControllerDidDismissPopover:]): Deleted.
(-[WKAirPlayRoutePicker _presentAirPlayPopoverAnimated:fromRect:]): Deleted.
(-[WKAirPlayRoutePicker _windowWillRotate:]): Deleted.
(-[WKAirPlayRoutePicker _windowDidRotate:]): Deleted.
(-[WKAirPlayRoutePicker _dismissAirPlayRoutePickerIPad]): Deleted.
(-[WKAirPlayRoutePicker showAirPlayPickerIPad:fromRect:]): Deleted.
(-[WKAirPlayRoutePicker showAirPlayPickerIPhone:]): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220209 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoImprove our support for referrer policies
cdumez@apple.com [Thu, 3 Aug 2017 17:19:44 +0000 (17:19 +0000)]
Improve our support for referrer policies
https://bugs.webkit.org/show_bug.cgi?id=175069
<rdar://problem/33677313>

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline several WPT tests now that more checks are passing.

* web-platform-tests/beacon/headers/header-referrer-origin-when-cross-origin-expected.txt:
* web-platform-tests/beacon/headers/header-referrer-same-origin-expected.txt:
* web-platform-tests/beacon/headers/header-referrer-strict-origin-when-cross-origin.https-expected.txt:
* web-platform-tests/beacon/headers/header-referrer-strict-origin.https-expected.txt:
* web-platform-tests/beacon/headers/header-referrer-unsafe-url.https-expected.txt:
* web-platform-tests/fetch/api/redirect/redirect-referrer-expected.txt:
* web-platform-tests/fetch/api/redirect/redirect-referrer-worker-expected.txt:
* web-platform-tests/fetch/api/request/request-init-001.sub-expected.txt:

Source/WebCore:

Improve our support for referrer policies. In particular, we now support the
additional following ones: "same-origin", "origin-when-cross-origin" and
"strict-origin-when-cross-origin".

This is as per the following specification:
- https://www.w3.org/TR/referrer-policy/#referrer-policies

Also refactor the code a bit for clarity: I merged the ReferrerPolicy enum and the
FetchOptions::ReferrerPolicy one.

Tests: http/tests/referrer-policy/origin-when-cross-origin/cross-origin-http-http.html
       http/tests/referrer-policy/origin-when-cross-origin/cross-origin-http.https.html
       http/tests/referrer-policy/origin-when-cross-origin/same-origin.html
       http/tests/referrer-policy/same-origin/cross-origin-http-http.html
       http/tests/referrer-policy/same-origin/cross-origin-http.https.html
       http/tests/referrer-policy/same-origin/same-origin.html
       http/tests/referrer-policy/strict-origin-when-cross-origin/cross-origin-http-http.html
       http/tests/referrer-policy/strict-origin-when-cross-origin/cross-origin-http.https.html
       http/tests/referrer-policy/strict-origin-when-cross-origin/same-origin.html
       http/tests/referrer-policy/strict-origin/cross-origin-http-http.html
       http/tests/referrer-policy/strict-origin/cross-origin-http.https.html
       http/tests/referrer-policy/strict-origin/same-origin.html

* Modules/fetch/FetchLoader.cpp:
(WebCore::FetchLoader::start):
* Modules/fetch/FetchReferrerPolicy.h:
* Modules/fetch/FetchReferrerPolicy.idl:
* Modules/fetch/FetchRequest.h:
* Modules/fetch/FetchRequestInit.h:
* dom/Document.cpp:
(WebCore::Document::processReferrerPolicy):
(WebCore::Document::applyQuickLookSandbox):
(WebCore::Document::applyContentDispositionAttachmentSandbox):
* dom/Document.h:
* loader/FetchOptions.h:
* loader/FrameNetworkingContext.h:
* loader/PingLoader.cpp:
(WebCore::PingLoader::sendBeacon):
Drop explicit call to SecurityPolicy::shouldHideReferrer(). This is already called inside
SecurityPolicy::generateReferrerHeader() and used only when needed, depending on the
actual referrer policy.

* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::updateHTTPRequestHeaders):
* loader/cache/CachedResourceRequest.cpp:
(WebCore::CachedResourceRequest::updateReferrerOriginAndUserAgentHeaders):
* page/SecurityPolicy.cpp:
(WebCore::referrerToOriginString):
(WebCore::SecurityPolicy::generateReferrerHeader):
* page/SecurityPolicy.h:
* platform/ReferrerPolicy.h:

Source/WebKit:

* WebProcess/Network/WebLoaderStrategy.cpp:
(WebKit::WebLoaderStrategy::loadResource):
(WebKit::WebLoaderStrategy::schedulePluginStreamLoad):

LayoutTests:

* http/tests/referrer-policy/origin-when-cross-origin/cross-origin-http-http-expected.txt: Added.
* http/tests/referrer-policy/origin-when-cross-origin/cross-origin-http-http.html: Added.
* http/tests/referrer-policy/origin-when-cross-origin/cross-origin-http.https-expected.txt: Added.
* http/tests/referrer-policy/origin-when-cross-origin/cross-origin-http.https.html: Added.
* http/tests/referrer-policy/origin-when-cross-origin/same-origin-expected.txt: Added.
* http/tests/referrer-policy/origin-when-cross-origin/same-origin.html: Added.
* http/tests/referrer-policy/resources/document.html: Added.
* http/tests/referrer-policy/same-origin/cross-origin-http-http-expected.txt: Added.
* http/tests/referrer-policy/same-origin/cross-origin-http-http.html: Added.
* http/tests/referrer-policy/same-origin/cross-origin-http.https-expected.txt: Added.
* http/tests/referrer-policy/same-origin/cross-origin-http.https.html: Added.
* http/tests/referrer-policy/same-origin/same-origin-expected.txt: Added.
* http/tests/referrer-policy/same-origin/same-origin.html: Added.
* http/tests/referrer-policy/strict-origin-when-cross-origin/cross-origin-http-http-expected.txt: Added.
* http/tests/referrer-policy/strict-origin-when-cross-origin/cross-origin-http-http.html: Added.
* http/tests/referrer-policy/strict-origin-when-cross-origin/cross-origin-http.https-expected.txt: Added.
* http/tests/referrer-policy/strict-origin-when-cross-origin/cross-origin-http.https.html: Added.
* http/tests/referrer-policy/strict-origin-when-cross-origin/same-origin-expected.txt: Added.
* http/tests/referrer-policy/strict-origin-when-cross-origin/same-origin.html: Added.
* http/tests/referrer-policy/strict-origin/cross-origin-http-http-expected.txt: Added.
* http/tests/referrer-policy/strict-origin/cross-origin-http-http.html: Added.
* http/tests/referrer-policy/strict-origin/cross-origin-http.https-expected.txt: Added.
* http/tests/referrer-policy/strict-origin/cross-origin-http.https.html: Added.
* http/tests/referrer-policy/strict-origin/same-origin-expected.txt: Added.
* http/tests/referrer-policy/strict-origin/same-origin.html: Added.
Add layout test coverage.

* http/tests/security/referrer-policy-invalid-expected.txt:
Rebaseline test now that console message has changed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220208 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoSupport ::marker pseudo-element
dbates@webkit.org [Thu, 3 Aug 2017 16:16:50 +0000 (16:16 +0000)]
Support ::marker pseudo-element
https://bugs.webkit.org/show_bug.cgi?id=141477

Reviewed by David Hyatt.

Source/WebCore:

Implements the ::marker pseudo element as per the CSS Pseudo-Element Module Level 4
spec., <https://drafts.csswg.org/css-pseudo-4> (Editor's Draft, 24 July 2017).

The ::marker pseudo element is a convenience pseudo element that allows a person to
style the appearance of a list item marker. For example, to render all list item
markers in bolded, blue text you would define a stylesheet with the following content:

li::marker {
    color: blue;
    font-weight: bold;
}

and this could be applied to a page that contains markup of the form:

<ol>
    <li>Item 1</li>
    <li>Item 2</li>
    ...
    <li>Item N-1</li>
    <li>Item N</li>
</ol>

Formerly to the achieve the same effect you would need to use a stylesheet of the form:

li {
    color: blue;
    font-weight: bold;
}

.list-item-content {
    all: initial;
}

and then write your markup to have the form:

<ol>
    <li><span class="list-item-content">Item 1</span></li>
    <li><span class="list-item-content">Item 2</span></li>
    ...
    <li><span class="list-item-content">Item N-1</span></li>
    <li><span class="list-item-content">Item N</span></li>
</ol>

The ::marker pseudo element only supports stylizing all font properties and the color property
of a list item marker.

Tests: fast/lists/list-marker-with-display.html
       http/wpt/css/css-pseudo-4/marker-and-other-pseudo-elements.html
       http/wpt/css/css-pseudo-4/marker-color.html
       http/wpt/css/css-pseudo-4/marker-font-properties.html
       http/wpt/css/css-pseudo-4/marker-inherit-values.html

* css/CSSSelector.cpp:
(WebCore::CSSSelector::pseudoId): Return the pseudo id for the ::marker pseudo element.
* css/CSSSelector.h: Add enumerator PseudoElementMarker to the pseudo element enum.
* css/RuleSet.cpp:
(WebCore::determinePropertyWhitelistType): Return whitelist type PropertyWhitelistMarker for ::marker
so that we match rules against the acceptable rules for ::marker.
* css/RuleSet.h: Add enumerator PropertyWhitelistMarker to the property whitelist type enum.
* css/SelectorPseudoElementTypeMap.in: Add "marker" to the list of pseudo element types.
* css/StyleResolver.cpp:
(WebCore::isValidMarkerStyleProperty): Determines if the specified CSS property is valid inside ::marker.
(WebCore::StyleResolver::CascadedProperties::addMatch): Only recognize CSS properties in the content block
of ::marker that match the ::marker whitelist policy.
* rendering/RenderListItem.cpp:
(WebCore::RenderListItem::computeMarkerStyle): Computes the style object for the list item marker. We
apply the user-agent style to the marker here as opposed to defining ::marker in the UA sheet as per
the spec. as an optimization to avoid having the style resolver apply the pseudo element to all elements.
For now, we always inherit style from the originating element (list item). Added FIXME to selectively
inherit styles.
(WebCore::RenderListItem::styleDidChange): Always apply the list marker style to the list marker renderer.
* rendering/RenderListItem.h:
* rendering/style/RenderStyleConstants.h: Add pseudo ID for the ::marker pseudo element.

LayoutTests:

Add tests that check we respect ::marker when rendering the list item marker. I will
submit all the tests in http/wpt/css/css-pseudo-4 to the Web Platform Tests repository
shortly and then import them into the WebKit repository in a subsequent commit.

* fast/lists/list-marker-with-display-expected.html: Added.
* fast/lists/list-marker-with-display.html: Added.
* http/wpt/css/css-pseudo-4/marker-and-other-pseudo-elements-expected.html: Added.
* http/wpt/css/css-pseudo-4/marker-and-other-pseudo-elements.html: Added.
* http/wpt/css/css-pseudo-4/marker-color-expected.html: Added.
* http/wpt/css/css-pseudo-4/marker-color.html: Added.
* http/wpt/css/css-pseudo-4/marker-font-properties-expected.html: Added.
* http/wpt/css/css-pseudo-4/marker-font-properties.html: Added.
* http/wpt/css/css-pseudo-4/marker-inherit-values-expected.html: Added.
* http/wpt/css/css-pseudo-4/marker-inherit-values.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220207 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] The test http/tests/security/contentSecurityPolicy/upgrade-insecure-requests...
pvollan@apple.com [Thu, 3 Aug 2017 16:01:54 +0000 (16:01 +0000)]
[Win] The test http/tests/security/contentSecurityPolicy/upgrade-insecure-requests/basic-upgrade.https.html is failing.
https://bugs.webkit.org/show_bug.cgi?id=159510

Reviewed by Daniel Bates.

Allow any https certificate when running tests.

* DumpRenderTree/win/DumpRenderTree.cpp:
(runTest):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220206 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] WebKit COM header file is not placed in the correct location.
pvollan@apple.com [Thu, 3 Aug 2017 14:47:48 +0000 (14:47 +0000)]
[Win] WebKit COM header file is not placed in the correct location.
https://bugs.webkit.org/show_bug.cgi?id=175101

Reviewed by Brent Fulgham.

After the transition to WebKitLegacy, the generated WebKit header files should still
be placed in the WebKit folder, so WebKit clients will find the header files in the
same place.

* WebKitLegacy.vcxproj/WebKitLegacy.proj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220205 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFactor common code in Style::*ChangeInvalidation into helper functions
antti@apple.com [Thu, 3 Aug 2017 14:45:29 +0000 (14:45 +0000)]
Factor common code in Style::*ChangeInvalidation into helper functions
https://bugs.webkit.org/show_bug.cgi?id=174312

Reviewed by Andreas Kling.

There is a lot of copy code here.

* Style/StyleInvalidationFunctions.h: Added.
(WebCore::Style::traverseRuleFeaturesInShadowTree):
(WebCore::Style::traverseRuleFeaturesForSlotted):
(WebCore::Style::traverseRuleFeatures):

    Add functions for traversing rule features that may affect style of an element.
    Use lambdas to implement client-specific behavior.

* WebCore.xcodeproj/project.pbxproj:
* style/AttributeChangeInvalidation.cpp:
(WebCore::Style::mayBeAffectedByAttributeChange):
(WebCore::Style::AttributeChangeInvalidation::invalidateStyle):
(WebCore::Style::mayBeAffectedByHostRules): Deleted.
(WebCore::Style::mayBeAffectedBySlottedRules): Deleted.
* style/ClassChangeInvalidation.cpp:
(WebCore::Style::ClassChangeInvalidation::invalidateStyle):
(WebCore::Style::mayBeAffectedByHostRules): Deleted.
(WebCore::Style::mayBeAffectedBySlottedRules): Deleted.
* style/IdChangeInvalidation.cpp:
(WebCore::Style::IdChangeInvalidation::invalidateStyle):
(WebCore::Style::mayBeAffectedByHostRules): Deleted.
(WebCore::Style::mayBeAffectedBySlottedRules): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220204 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EME] CDM constructor assigns CDMPrivate member multiple times
zandobersek@gmail.com [Thu, 3 Aug 2017 14:40:47 +0000 (14:40 +0000)]
[EME] CDM constructor assigns CDMPrivate member multiple times
https://bugs.webkit.org/show_bug.cgi?id=175128

Reviewed by Xabier Rodriguez-Calvar.

In the CDM class constructor, iterate over the registered CDM
factories, finding one that supports the specified key system.
A CDMPrivate object is created through that factory, and the
iteration is now stopped at that point, while previously it
contined to potentially create CDMPrivate objects through
other factories.

Helper createCDMPrivateForKeySystem() function is removed.

* Modules/encryptedmedia/CDM.cpp:
(WebCore::CDM::CDM):
(WebCore::createCDMPrivateForKeySystem): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220203 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoDon't always recalc the style of display: contents elements.
commit-queue@webkit.org [Thu, 3 Aug 2017 13:20:30 +0000 (13:20 +0000)]
Don't always recalc the style of display: contents elements.
https://bugs.webkit.org/show_bug.cgi?id=172753

Patch by Emilio Cobos Ãlvarez <ecobos@igalia.com> on 2017-08-03
Reviewed by Antti Koivisto.

No new tests (no functionality change). This only removes an
inefficiency.

* dom/Element.cpp:
(WebCore::Element::existingComputedStyle):
* dom/Element.h:
* style/RenderTreeUpdater.cpp:
(WebCore::RenderTreeUpdater::updateRenderTree):
(WebCore::RenderTreeUpdater::updateElementRenderer):
* style/StyleTreeResolver.cpp:
(WebCore::Style::renderOrDisplayContentsStyle):
(WebCore::Style::TreeResolver::resolveElement):
(WebCore::Style::TreeResolver::createAnimatedElementUpdate):
(WebCore::Style::shouldResolveElement):
(WebCore::Style::TreeResolver::resolveComposedTree):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220202 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Linux][WTF] Use one global semaphore to notify thread suspend and resume completion
utatane.tea@gmail.com [Thu, 3 Aug 2017 11:17:44 +0000 (11:17 +0000)]
[Linux][WTF] Use one global semaphore to notify thread suspend and resume completion
https://bugs.webkit.org/show_bug.cgi?id=175124

Reviewed by Carlos Garcia Campos.

POSIX sem_t is used to notify thread suspend and resume completion in Linux ports
since sem_post is async-signal-safe function. Since we guard suspend() and resume()
with one global lock, this semaphore is also guarded by this lock. So we do not need
to have semaphore per WTF::Thread.

This patch introduces one global Semaphore. And drop per thread semaphore.

* wtf/Threading.h:
* wtf/ThreadingPthreads.cpp:
(WTF::Thread::~Thread):
(WTF::Semaphore::Semaphore):
(WTF::Semaphore::~Semaphore):
(WTF::Semaphore::wait):
(WTF::Semaphore::post):
(WTF::Thread::signalHandlerSuspendResume):
(WTF::Thread::initializePlatformThreading):
(WTF::Thread::suspend):
(WTF::Thread::resume):
(WTF::Thread::Thread): Deleted.
* wtf/ThreadingWin.cpp:
(WTF::Thread::Thread): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220201 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Remove obsolete failure expectation for narrow-non-breaking-space.html.
commit-queue@webkit.org [Thu, 3 Aug 2017 10:03:24 +0000 (10:03 +0000)]
[GTK] Remove obsolete failure expectation for narrow-non-breaking-space.html.
https://bugs.webkit.org/show_bug.cgi?id=139493

Unreviewed test gardening.

It likely started passing in r205826 due to its change in Font.cpp.

Patch by Ms2ger <Ms2ger@igalia.com> on 2017-08-03

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220200 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Skip timezone-dependent Sputnik tests.
commit-queue@webkit.org [Thu, 3 Aug 2017 09:33:31 +0000 (09:33 +0000)]
[GTK] Skip timezone-dependent Sputnik tests.
https://bugs.webkit.org/show_bug.cgi?id=175120

Unreviewed test gardening.

These tests are skipped in the platform-neutral TestExpectations file because
they only pass in Pacific Time (see bug 42625). There doesn't seem to be a
reason for gtk to run them, and they do pass on the buildbot.

Patch by Ms2ger <Ms2ger@igalia.com> on 2017-08-03

* platform/gtk/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220199 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Linux][WTF] Reduce sizeof(WTF::Thread) by using a pointer to PlatformRegisters
utatane.tea@gmail.com [Thu, 3 Aug 2017 08:53:00 +0000 (08:53 +0000)]
[Linux][WTF] Reduce sizeof(WTF::Thread) by using a pointer to PlatformRegisters
https://bugs.webkit.org/show_bug.cgi?id=175119

Reviewed by Carlos Garcia Campos.

sizeof(PlatformRegisters) is so large. In my Linux box, it is 256. It enlarges the sizeof(WTF::Thread).
However, it is not necessary to hold it in WTF::Thread member. Thread's ucontext data and its stack is
effective while suspending the thread. So, we can just use the pointer to the PlatformRegister instead
of copying it to the member of the WTF::Thread.

* wtf/Threading.h:
* wtf/ThreadingPthreads.cpp:
(WTF::Thread::signalHandlerSuspendResume):
(WTF::Thread::getRegisters):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220198 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: add stack trace information for each RecordingAction
drousso@apple.com [Thu, 3 Aug 2017 06:18:08 +0000 (06:18 +0000)]
Web Inspector: add stack trace information for each RecordingAction
https://bugs.webkit.org/show_bug.cgi?id=174663

Reviewed by Joseph Pecoraro.

Source/JavaScriptCore:

* inspector/ScriptCallFrame.h:
Add `operator==` so that when a ScriptCallFrame object is held in a Vector, calling `find`
with an existing value doesn't need require a functor and can use existing code.

* interpreter/StackVisitor.h:
* interpreter/StackVisitor.cpp:
(JSC::StackVisitor::Frame::isWasmFrame const): Inlined in header.

Source/WebCore:

Tests: inspector/canvas/recording-2d.html
       inspector/model/recording.html

* inspector/InspectorCanvas.h:
* inspector/InspectorCanvas.cpp:
(WebCore::InspectorCanvas::indexForData):
(WebCore::InspectorCanvas::buildAction):

Source/WebInspectorUI:

* Localizations/en.lproj/localizedStrings.js:
* UserInterface/Main.html:

* UserInterface/Models/Recording.js:
(WI.Recording.prototype.swizzle):
Add Array type for swizzling array values.

* UserInterface/Models/RecordingAction.js:
(WI.RecordingAction):
(WI.RecordingAction.fromPayload):
(WI.RecordingAction.prototype.get trace):
(WI.RecordingAction.prototype.swizzle):
(WI.RecordingAction.prototype.toJSON):

* UserInterface/Views/RecordingTraceDetailsSidebarPanel.js: Added.
(WI.RecordingTraceDetailsSidebarPanel):
(WI.RecordingTraceDetailsSidebarPanel.disallowInstanceForClass):
(WI.RecordingTraceDetailsSidebarPanel.prototype.inspect):
(WI.RecordingTraceDetailsSidebarPanel.prototype.set recording):
(WI.RecordingTraceDetailsSidebarPanel.prototype.updateActionIndex):
* UserInterface/Views/RecordingTraceDetailsSidebarPanel.css: Added.
(.sidebar > .panel.details.recording-trace > .content > .call-frame):
(.sidebar > .details.recording-trace > .content > .no-trace-data):
(.sidebar > .details.recording-trace > .content > .no-trace-data > .message):

* UserInterface/Views/RecordingTabContentView.js:
(WI.RecordingTabContentView):

* UserInterface/Views/RecordingActionTreeElement.js:
(WI.RecordingActionTreeElement.prototype.populateContextMenu):

* UserInterface/Views/CallFrameView.css:
(.call-frame):
(body[dir=ltr] .call-frame .icon):
(body[dir=rtl] .call-frame .icon):
Apply the same trailing margin for CallFrameView icons as TreeElement.

LayoutTests:

* inspector/canvas/recording-2d-expected.txt:
* inspector/canvas/recording-2d.html:
* inspector/model/recording-expected.txt:
* inspector/model/recording.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220188 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, build fix for Windows port
utatane.tea@gmail.com [Thu, 3 Aug 2017 06:16:38 +0000 (06:16 +0000)]
Unreviewed, build fix for Windows port
https://bugs.webkit.org/show_bug.cgi?id=174716

This ugliness will be fixed in https://bugs.webkit.org/show_bug.cgi?id=175013.

* wtf/ThreadHolder.h:
* wtf/Threading.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220187 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMerge WTFThreadData to Thread::current
utatane.tea@gmail.com [Thu, 3 Aug 2017 06:03:18 +0000 (06:03 +0000)]
Merge WTFThreadData to Thread::current
https://bugs.webkit.org/show_bug.cgi?id=174716

Reviewed by Mark Lam.

Source/JavaScriptCore:

Use Thread::current() instead.

* API/JSContext.mm:
(+[JSContext currentContext]):
(+[JSContext currentThis]):
(+[JSContext currentCallee]):
(+[JSContext currentArguments]):
(-[JSContext beginCallbackWithData:calleeValue:thisValue:argumentCount:arguments:]):
(-[JSContext endCallbackWithData:]):
* heap/Heap.cpp:
(JSC::Heap::requestCollection):
* runtime/Completion.cpp:
(JSC::checkSyntax):
(JSC::checkModuleSyntax):
(JSC::evaluate):
(JSC::loadAndEvaluateModule):
(JSC::loadModule):
(JSC::linkAndEvaluateModule):
(JSC::importModule):
* runtime/Identifier.cpp:
(JSC::Identifier::checkCurrentAtomicStringTable):
* runtime/InitializeThreading.cpp:
(JSC::initializeThreading):
* runtime/JSLock.cpp:
(JSC::JSLock::didAcquireLock):
(JSC::JSLock::willReleaseLock):
(JSC::JSLock::dropAllLocks):
(JSC::JSLock::grabAllLocks):
* runtime/JSLock.h:
* runtime/VM.cpp:
(JSC::VM::VM):
(JSC::VM::updateStackLimits):
(JSC::VM::committedStackByteCount):
* runtime/VM.h:
(JSC::VM::isSafeToRecurse const):
* runtime/VMEntryScope.cpp:
(JSC::VMEntryScope::VMEntryScope):
* runtime/VMInlines.h:
(JSC::VM::ensureStackCapacityFor):
* yarr/YarrPattern.cpp:
(JSC::Yarr::YarrPatternConstructor::isSafeToRecurse const):

Source/WebCore:

Use Thread::current() instead.

* fileapi/AsyncFileStream.cpp:
* platform/ThreadGlobalData.cpp:
(WebCore::ThreadGlobalData::ThreadGlobalData):
* platform/graphics/cocoa/WebCoreDecompressionSession.h:
* platform/ios/wak/WebCoreThread.mm:
(StartWebThread):
* workers/WorkerThread.cpp:
(WebCore::WorkerThread::workerThread):

Source/WTF:

We placed thread specific data in WTFThreadData previously. But now, we have a new good place
to put thread specific data: WTF::Thread. Before this patch, WTFThreadData and WTF::Thread
sometimes have the completely same fields (m_stack etc.) due to initialization order limitations.
This patch merges WTFThreadData to WTF::Thread. We apply WTFThreadData's initialization style
to WTF::Thread. So, WTF::Thread's holder now uses fast TLS for darwin environment. Thus,
Thread::current() access is now accelerated. And WTF::Thread::current() can be accessed even
before calling WTF::initializeThreading.

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
* wtf/LockAlgorithm.h:
* wtf/LockAlgorithmInlines.h:
* wtf/MainThread.h:
* wtf/ParkingLot.cpp:
* wtf/StackStats.cpp:
(WTF::StackStats::PerThreadStats::PerThreadStats):
(WTF::StackStats::CheckPoint::CheckPoint):
(WTF::StackStats::CheckPoint::~CheckPoint):
(WTF::StackStats::probe):
(WTF::StackStats::LayoutCheckPoint::LayoutCheckPoint):
* wtf/ThreadHolder.cpp:
(WTF::ThreadHolder::initializeCurrent):
* wtf/ThreadHolder.h:
(WTF::ThreadHolder::ThreadHolder):
(WTF::ThreadHolder::currentMayBeNull):
(WTF::ThreadHolder::current):
* wtf/ThreadHolderPthreads.cpp:
(WTF::ThreadHolder::initializeKey):
(WTF::ThreadHolder::initialize):
(WTF::ThreadHolder::destruct):
(WTF::ThreadHolder::initializeOnce): Deleted.
(WTF::ThreadHolder::current): Deleted.
* wtf/ThreadHolderWin.cpp:
(WTF::ThreadHolder::initializeKey):
(WTF::ThreadHolder::currentDying):
(WTF::ThreadHolder::initialize):
(WTF::ThreadHolder::initializeOnce): Deleted.
(WTF::ThreadHolder::current): Deleted.
* wtf/Threading.cpp:
(WTF::Thread::initializeInThread):
(WTF::Thread::entryPoint):
(WTF::Thread::create):
(WTF::Thread::didExit):
(WTF::initializeThreading):
(WTF::Thread::currentMayBeNull): Deleted.
* wtf/Threading.h:
(WTF::Thread::current):
(WTF::Thread::atomicStringTable):
(WTF::Thread::setCurrentAtomicStringTable):
(WTF::Thread::stackStats):
(WTF::Thread::savedStackPointerAtVMEntry):
(WTF::Thread::setSavedStackPointerAtVMEntry):
(WTF::Thread::savedLastStackTop):
(WTF::Thread::setSavedLastStackTop):
* wtf/ThreadingPrimitives.h:
* wtf/ThreadingPthreads.cpp:
(WTF::Thread::createCurrentThread):
(WTF::Thread::current): Deleted.
* wtf/ThreadingWin.cpp:
(WTF::Thread::createCurrentThread):
(WTF::Thread::current): Deleted.
* wtf/WTFThreadData.cpp: Removed.
* wtf/WTFThreadData.h: Removed.
* wtf/text/AtomicString.cpp:
* wtf/text/AtomicStringImpl.cpp:
(WTF::stringTable):
* wtf/text/AtomicStringTable.cpp:
(WTF::AtomicStringTable::create):
* wtf/text/AtomicStringTable.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220186 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoNetworkResourceLoader::setDefersLoading() may cause start() to be called multiple...
cdumez@apple.com [Thu, 3 Aug 2017 05:57:45 +0000 (05:57 +0000)]
NetworkResourceLoader::setDefersLoading() may cause start() to be called multiple times
https://bugs.webkit.org/show_bug.cgi?id=175109
<rdar://problem/33363169>

Reviewed by Brady Eidson.

Source/WebKit:

If NetworkResourceLoader::setDefersLoading(false) is called by the client while m_networkLoad is null
then we call NetworkResourceLoader::start() to start the load. This is needed in the case where
a NetworkResourceLoader is constructed in deferred mode so that the load can later be started via
setDefersLoading(). However, it is possible for setDefersLoading(false) to be called when start()
has already been called, which causes start() to be called multiple times and leads to an assertion
hit in debug.

Normally, setDefersLoading(false) returns without calling start() if m_networkLoad is not null. It
relies on m_networkLoad being non-null to determine if start() should be called. This is bad because
start() checks the disk cache asynchronously *before* calling startNetworkLoad() and initializing
m_networkLoad. Therefore, if setDefersLoading(false) is called *while* we are checking the cache,
then we will call incorrectly call start() again.

In the case of the radar, this happens when we:
1. Call start() and check the disk cache
2. Retrieve a cached redirect from the cache, which causes a WillSendRequest IPC to be sent to the
   WebProcess
3. The WebProcess calls setDefersLoading(true), sends the ContinueWillSendRequest IPC back and
   then calls setDefersLoading(false)

The call to continueWillSendRequest() causes us to retrieve the redirected entry from the cache,
asynchronously, which will return no entry and start a load.
The later call to setDefersLoading(false) causes us to call start() again and we end up back to step 1.

* NetworkProcess/NetworkResourceLoader.cpp:
(WebKit::NetworkResourceLoader::start):
(WebKit::NetworkResourceLoader::setDefersLoading):
* NetworkProcess/NetworkResourceLoader.h:

LayoutTests:

Extend test coverage to cover cacheable redirects to a resource that needs
revalidation, similarly to the case in the radar.

* http/tests/cache/disk-cache/disk-cache-redirect-expected.txt:
* http/tests/cache/disk-cache/disk-cache-redirect.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220185 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoLLInt should do pointer caging
fpizlo@apple.com [Thu, 3 Aug 2017 05:48:59 +0000 (05:48 +0000)]
LLInt should do pointer caging
https://bugs.webkit.org/show_bug.cgi?id=175036

Reviewed by Keith Miller.

Implementing this in the LLInt was challenging because offlineasm did not previously know
how to load from globals. This teaches it how to do that on Darwin/x86_64, which happens
to be where the Gigacage is enabled right now.

* llint/LLIntOfflineAsmConfig.h:
* llint/LowLevelInterpreter64.asm:
* offlineasm/ast.rb:
* offlineasm/x86.rb:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220184 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoNeverDestroyed related leaks seen on bots
commit-queue@webkit.org [Thu, 3 Aug 2017 04:46:46 +0000 (04:46 +0000)]
NeverDestroyed related leaks seen on bots
https://bugs.webkit.org/show_bug.cgi?id=175113

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-08-02
Reviewed by Yusuke Suzuki.

* wtf/NeverDestroyed.h:
(WTF::NeverDestroyed::NeverDestroyed):
Previously the result of makeNeverDestroyed was not always moving into
the `static NeverDestroyed` static local variable. In some cases it would
re-invoke the constructor, creating a new NeverDestroyed object. In the
case of a Vector it was causing leaks.

Adding a move constructor convinces the compiler to move the result
of makeNeverDestroyed into the NeverDestroyed static. It doesn't actually
invoke the move constructor here, which I believe means it is deciding
to perform optional copy elision optimization.
'http://en.cppreference.com/w/cpp/language/copy_elision

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220183 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: add TreeElement virtualization for the Recording tab
drousso@apple.com [Thu, 3 Aug 2017 04:21:32 +0000 (04:21 +0000)]
Web Inspector: add TreeElement virtualization for the Recording tab
https://bugs.webkit.org/show_bug.cgi?id=174968

Reviewed by Joseph Pecoraro.

* UserInterface/Views/RecordingNavigationSidebarPanel.js:
(WI.RecordingNavigationSidebarPanel):

* UserInterface/Views/TreeOutline.js:
(WI.TreeOutline):
(WI.TreeOutline.prototype.get virtualized):
(WI.TreeOutline.prototype.registerScrollVirtualizer):
(WI.TreeOutline.prototype.updateVirtualizedElements.walk):
(WI.TreeOutline.prototype.updateVirtualizedElements):
Add spacer elements before and after the TreeOutline element that will size to ensure that
the TreeOutline node still takes up the same amount of space after some of the TreeElements
are removed. Whenever the scroll of the container view changes, recalculate the visible area
and add/remove TreeElements based on whether they would be in that. This is only possible if
every TreeElement has the same vertical height, which is given when setting up the scroll
listener on the container view.

* UserInterface/Views/TreeElement.js:
(WI.TreeElement.prototype.set hidden):
(WI.TreeElement.prototype._attach):
(WI.TreeElement.prototype.collapse):
(WI.TreeElement.prototype.expand):
(WI.TreeElement.prototype.reveal):
If the TreeOutline is being virtualized, don't add each TreeElement's node to the DOM. They
will be added at the end of the frame (via setTimeout) if they are within the visible + padding
area of the TreeOutline.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220180 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoHTTP tests with 'https' suffix are only run over HTTPS for WK2, not WK1
commit-queue@webkit.org [Thu, 3 Aug 2017 04:19:10 +0000 (04:19 +0000)]
HTTP tests with 'https' suffix are only run over HTTPS for WK2, not WK1
https://bugs.webkit.org/show_bug.cgi?id=175089

Patch by Youenn Fablet <youenn@apple.com> on 2017-08-02
Reviewed by Chris Dumez.

Tools:

* DumpRenderTree/TestOptions.mm:
(TestOptions::TestOptions): Using absolutePath if available.
* DumpRenderTree/mac/DumpRenderTree.mm:
(computeTestURL): Removing http/tests/ specific URL computation.
* Scripts/webkitpy/port/driver.py:
(Driver._command_from_driver_input): Making webkitpy passing HTTP urls for HTTP served tests to all test runners, including WK1.

LayoutTests:

Removing no longer needed expectations.

* platform/ios-wk1/http/tests/security/contentSecurityPolicy/upgrade-insecure-requests/iframe-upgrade.https-expected.txt: Removed.
* platform/mac-wk1/http/tests/security/contentSecurityPolicy/upgrade-insecure-requests/iframe-upgrade.https-expected.txt: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220179 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoNetworkRTCProvider::createResolver() leaks CFHost objects
akling@apple.com [Thu, 3 Aug 2017 03:53:05 +0000 (03:53 +0000)]
NetworkRTCProvider::createResolver() leaks CFHost objects
https://bugs.webkit.org/show_bug.cgi?id=175103
<rdar://problem/33690347>

Reviewed by Youenn Fablet.

Add a missing adoptCF() so we don't leak the CFHost.

* NetworkProcess/webrtc/NetworkRTCProvider.cpp:
(WebKit::NetworkRTCProvider::createResolver):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220178 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix crashes in GC creating a document fragment on a background thread
commit-queue@webkit.org [Thu, 3 Aug 2017 03:49:53 +0000 (03:49 +0000)]
Fix crashes in GC creating a document fragment on a background thread
https://bugs.webkit.org/show_bug.cgi?id=175111

Patch by Sam Weinig <sam@webkit.org> on 2017-08-02
Reviewed by Chris Dumez.

r220095 (https://webkit.org/b/175006) change JSHTMLTemplateElement from using a
private name + property to manager the lifetime of the reference DocumentFragment
to using the idiomatic visitAdditionalChildren. Unfortunately, the function to access
the DocumentFragment lazily creates it. If this lazy creation happens on a GC thread,
badness ensues. This introduces an accessor that returns the DocumentFragment if it
has been created or null if it has not.

* bindings/js/JSHTMLTemplateElementCustom.cpp:
(WebCore::JSHTMLTemplateElement::visitAdditionalChildren):
* html/HTMLTemplateElement.cpp:
(WebCore::HTMLTemplateElement::contentIfAvailable):
* html/HTMLTemplateElement.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220177 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[WebIDL] Simplify [EnabledBySettings] extended attribute code to not require passing...
commit-queue@webkit.org [Thu, 3 Aug 2017 03:25:30 +0000 (03:25 +0000)]
[WebIDL] Simplify [EnabledBySettings] extended attribute code to not require passing a global object to finishCreation
https://bugs.webkit.org/show_bug.cgi?id=175087

Patch by Sam Weinig <sam@webkit.org> on 2017-08-02
Reviewed by Chris Dumez.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateImplementation):
(GeneratePrototypeDeclaration):
Remove unnecessary passing of the global object to finishCreation for [EnabledBySettings].

* bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:
* bindings/scripts/test/JS/JSTestNode.cpp:
* bindings/scripts/test/JS/JSTestObj.cpp:
Update tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220176 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoSweeping should only scribble when sweeping to free list
fpizlo@apple.com [Thu, 3 Aug 2017 01:57:37 +0000 (01:57 +0000)]
Sweeping should only scribble when sweeping to free list
https://bugs.webkit.org/show_bug.cgi?id=175105

Reviewed by Saam Barati.

I just saw a crash on the bots where a destructor call attempt dereferenced scribbled memory. This
can happen because the bump path of specializedSweep will scribble in SweepOnly, which replaces the
zap word (i.e. 0) with the scribble word (i.e. 0xbadbeef0). This is a recent regression, since we
didn't used to do destruction on the bump path. No destruction, no zapping. Looking at the pop
path, we only scribble when we SweepToFreeList. This ensures that we only overwrite the zap word
when it doesn't matter anyway because we're building a free list.

This is a fix for those crashes on the bots because it means that we'll no longer scribble over the
zap.

* heap/MarkedBlockInlines.h:
(JSC::MarkedBlock::Handle::specializedSweep):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220175 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMarked http/tests/appcache/deferred-events-delete-while-raising-timer.html as flaky.
jlewis3@apple.com [Thu, 3 Aug 2017 01:55:59 +0000 (01:55 +0000)]
Marked http/tests/appcache/deferred-events-delete-while-raising-timer.html as flaky.
https://bugs.webkit.org/show_bug.cgi?id=175107

Unreviewed test gardening.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220174 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Linux] JSTests/wasm/stress/oom.js should not run on Linux
clopez@igalia.com [Thu, 3 Aug 2017 01:42:13 +0000 (01:42 +0000)]
[Linux] JSTests/wasm/stress/oom.js should not run on Linux
https://bugs.webkit.org/show_bug.cgi?id=175100

Reviewed by Mark Lam.

The JSC test JSTests/wasm/stress/oom.js tries to use all the
available memory until an out of memory exception happens.

The Linux kernel is more tuned for server workloads than for GUI
responsiveness. When a process tries to use a lot of memory, Linux
will do its best to serve the request. This usually translates to
free physical RAM by writing to disk dirty pages and/or moving out
less recently used pages to swap (disk storage).
Meanwhile it does this, the system will become unresponsive and this
leads to freezes that can last even some minutes on the worst cases.

Therefore, let's skip this test on Linux as it will cause more harm
than good on the Linux bots or on the machines of Linux developers.

* wasm/stress/oom.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220173 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agogit-svn-id: https://svn.webkit.org/repository/webkit/trunk@220166 268f45cc-cd09-0410...
lforschler@apple.com [Thu, 3 Aug 2017 01:32:27 +0000 (01:32 +0000)]
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220166 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAll C++ accesses to JSObject::m_butterfly should do caging
fpizlo@apple.com [Thu, 3 Aug 2017 01:32:07 +0000 (01:32 +0000)]
All C++ accesses to JSObject::m_butterfly should do caging
https://bugs.webkit.org/show_bug.cgi?id=175039

Reviewed by Keith Miller.

Source/JavaScriptCore:

Makes JSObject::m_butterfly a AuxiliaryBarrier<CagedPtr<Butterfly>> and adopts the CagedPtr<> API.
This ensures that you can't cause C++ code to access a butterfly that has been rewired to point
outside the gigacage.

* runtime/JSArray.cpp:
(JSC::JSArray::setLength):
(JSC::JSArray::pop):
(JSC::JSArray::push):
(JSC::JSArray::shiftCountWithAnyIndexingType):
(JSC::JSArray::unshiftCountWithAnyIndexingType):
(JSC::JSArray::fillArgList):
(JSC::JSArray::copyToArguments):
* runtime/JSObject.cpp:
(JSC::JSObject::heapSnapshot):
(JSC::JSObject::createInitialIndexedStorage):
(JSC::JSObject::createArrayStorage):
(JSC::JSObject::convertUndecidedToInt32):
(JSC::JSObject::convertUndecidedToDouble):
(JSC::JSObject::convertUndecidedToContiguous):
(JSC::JSObject::convertInt32ToDouble):
(JSC::JSObject::convertInt32ToArrayStorage):
(JSC::JSObject::convertDoubleToContiguous):
(JSC::JSObject::convertDoubleToArrayStorage):
(JSC::JSObject::convertContiguousToArrayStorage):
(JSC::JSObject::defineOwnIndexedProperty):
(JSC::JSObject::putByIndexBeyondVectorLengthWithoutAttributes):
(JSC::JSObject::ensureLengthSlow):
(JSC::JSObject::allocateMoreOutOfLineStorage):
* runtime/JSObject.h:
(JSC::JSObject::canGetIndexQuickly):
(JSC::JSObject::getIndexQuickly):
(JSC::JSObject::tryGetIndexQuickly const):
(JSC::JSObject::canSetIndexQuickly):
(JSC::JSObject::setIndexQuickly):
(JSC::JSObject::initializeIndex):
(JSC::JSObject::initializeIndexWithoutBarrier):
(JSC::JSObject::butterfly const):
(JSC::JSObject::butterfly):

Source/WTF:

Adds a smart pointer class that does various kinds of caging for you.

* WTF.xcodeproj/project.pbxproj:
* wtf/CMakeLists.txt:
* wtf/CagedPtr.h: Added.
(WTF::CagedPtr::CagedPtr):
(WTF::CagedPtr::get const):
(WTF::CagedPtr::getMayBeNull const):
(WTF::CagedPtr::operator== const):
(WTF::CagedPtr::operator!= const):
(WTF::CagedPtr::operator bool const):
(WTF::CagedPtr::operator* const):
(WTF::CagedPtr::operator-> const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220165 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemoved bad expectations and marked test as flaky.
jlewis3@apple.com [Thu, 3 Aug 2017 00:54:58 +0000 (00:54 +0000)]
Removed bad expectations and marked test as flaky.
https://bugs.webkit.org/show_bug.cgi?id=175061

Unreviewed test gardening.

* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/basic/mode-no-cors-worker-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic.any-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic.any.worker-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-no-preflight.any-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-no-preflight.any.worker-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin.any-expected.txt: Removed.
* platform/mac-elcapitan-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin.any.worker-expected.txt: Removed.
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220164 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION (r207155): Unable to switch sheets when previewing Numbers '09 spreadsheets
aestes@apple.com [Thu, 3 Aug 2017 00:46:31 +0000 (00:46 +0000)]
REGRESSION (r207155): Unable to switch sheets when previewing Numbers '09 spreadsheets
https://bugs.webkit.org/show_bug.cgi?id=175098
<rdar://problem/31416763>

Reviewed by Daniel Bates.

Source/WebCore:

r207155 enabled sandboxing on the frame displaying a QuickLook preview. This restricted
frames within the sandbox from navigating their sandboxed siblings or ancestors, which
breaks the functionality of multi-sheet Numbers '09 spreadsheet previews. These previews
contain a frameset with a table of contents frame and a content frame, and the table of
contents frame needs to be able to navigate the content frame when the sheet selection
changes.

Fix this by disabling the SandboxNavigation flag in the QuickLook sandbox. Frames within the
sandbox will be able to navigate each other, but will not be able to navigate the top frame
(due to SandboxTopNavigation still being enabled), nor will they be able to navigate any
other ancestor frame outside the sandbox (due to QuickLook previews being in a different
origin than the hosting frame). These two cases are covered by existing tests.

Test: quicklook/multi-sheet-numbers-09.html

* dom/Document.cpp:
(WebCore::Document::applyQuickLookSandbox): Added a call to
disableSandboxFlags(SandboxNavigation) after applying the content security policy.
* dom/SecurityContext.h:
(WebCore::SecurityContext::disableSandboxFlags): Defined disableSandboxFlags().

LayoutTests:

* quicklook/multi-sheet-numbers-09-expected.txt: Added.
* quicklook/multi-sheet-numbers-09.html: Added.
* quicklook/resources/multi-sheet-numbers-09.numbers: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220163 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, skip newly imported WPT that is slow in Debug builds.
cdumez@apple.com [Wed, 2 Aug 2017 23:56:43 +0000 (23:56 +0000)]
Unreviewed, skip newly imported WPT that is slow in Debug builds.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220160 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoinspector/runtime/CommandLineAPI-inspect.html is flaky.
commit-queue@webkit.org [Wed, 2 Aug 2017 22:46:27 +0000 (22:46 +0000)]
inspector/runtime/CommandLineAPI-inspect.html is flaky.
https://bugs.webkit.org/show_bug.cgi?id=175092

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2017-08-02
Reviewed by Brian Burg.

* inspector/runtime/CommandLineAPI-inspect-expected.txt:
* inspector/runtime/CommandLineAPI-inspect.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220159 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agocheck-webkit-style: deleting lines in a file runs the linter on the whole file
jbedard@apple.com [Wed, 2 Aug 2017 22:35:44 +0000 (22:35 +0000)]
check-webkit-style: deleting lines in a file runs the linter on the whole file
https://bugs.webkit.org/show_bug.cgi?id=175078

Reviewed by David Kilzer.

Deleting lines in a file should not cause linter errors to be blamed on the patch.
<https://bugs.webkit.org/show_bug.cgi?id=86142> is an example of this happening.

* Scripts/webkitpy/style/checkers/test_expectations.py:
(TestExpectationsChecker._should_log_linter_warning): Do not log a linter error if the file it is associated with only has deleted lines
* Scripts/webkitpy/style/main_unittest.py:
(ExpectationLinterInStyleCheckerTest.test_linter_duplicate_line): Added files should have every line number in the file when processing.
(ExpectationLinterInStyleCheckerTest.test_linter_duplicate_line_only_deletes): Test case where the file with the linter errors only contained deletes.
(ExpectationLinterInStyleCheckerTest.test_linter_added_file_with_error): Added files should have every line number in the file when processing.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220158 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoVersioning.
jmarcell@apple.com [Wed, 2 Aug 2017 22:23:39 +0000 (22:23 +0000)]
Versioning.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220155 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoIf Gigacage is disabled, bmalloc should service large aligned memory allocation reque...
fpizlo@apple.com [Wed, 2 Aug 2017 21:50:33 +0000 (21:50 +0000)]
If Gigacage is disabled, bmalloc should service large aligned memory allocation requests through vmAllocate
https://bugs.webkit.org/show_bug.cgi?id=175085

Reviewed by Saam Barati.

This fixes a problem where if we used gmalloc, WebAssembly memory allocations would still use
bmalloc's large allocator.

We want to use the page allocator for those "large" allocations when the Gigacage is disabled.

* bmalloc/DebugHeap.cpp:
(bmalloc::DebugHeap::DebugHeap):
(bmalloc::DebugHeap::memalignLarge):
(bmalloc::DebugHeap::freeLarge):
* bmalloc/DebugHeap.h:
* bmalloc/Heap.cpp:
(bmalloc::Heap::tryAllocateLarge):
(bmalloc::Heap::deallocateLarge):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220154 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[MSE] Removing samples when presentation order does not match decode order can cause...
jer.noble@apple.com [Wed, 2 Aug 2017 21:50:13 +0000 (21:50 +0000)]
[MSE] Removing samples when presentation order does not match decode order can cause bad behavior.
https://bugs.webkit.org/show_bug.cgi?id=175091

Reviewed by Eric Carlson.

Address follow-up comments to r219519.

* Modules/mediasource/SourceBuffer.cpp:
(WebCore::SourceBuffer::removeCodedFrames):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220153 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove unused and obsolete setting mediaDocumentEntersFullscreenAutomatically
commit-queue@webkit.org [Wed, 2 Aug 2017 20:45:42 +0000 (20:45 +0000)]
Remove unused and obsolete setting mediaDocumentEntersFullscreenAutomatically
https://bugs.webkit.org/show_bug.cgi?id=175080

Patch by Jeremy Jones <jeremyj@apple.com> on 2017-08-02
Reviewed by Jon Lee.

Source/WebCore:

No new tests because this only removes unused code.

This is obsolete because of the alternate solution in
https://bugs.webkit.org/show_bug.cgi?id=174850

* page/Settings.in:

Source/WebKit:

This is obsolete because of the alternate solution in
https://bugs.webkit.org/show_bug.cgi?id=174850

* Shared/WebPreferencesDefinitions.h:
* UIProcess/API/Cocoa/WKPreferences.mm:
(-[WKPreferences _setMediaDocumentEntersFullscreenAutomatically:]): Deleted.
(-[WKPreferences _mediaDocumentEntersFullscreenAutomatically]): Deleted.
* UIProcess/API/Cocoa/WKPreferencesPrivate.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220151 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agowebkitpy: Allow caller to specify response to unicode encode/decode error in filesystem
jbedard@apple.com [Wed, 2 Aug 2017 20:01:32 +0000 (20:01 +0000)]
webkitpy: Allow caller to specify response to unicode encode/decode error in filesystem
https://bugs.webkit.org/show_bug.cgi?id=175075

Reviewed by David Kilzer.

We have no way of handling text files with illegal unicode characters. Allow the callers of
filesystem.read_text_file to specify one of Python 2.7's supported responses ('strict', 'ignore', or
'replace'). See https://docs.python.org/2/howto/unicode.html for details on these responses.

* Scripts/webkitpy/common/system/filesystem.py:
(FileSystem.read_text_file): Allow caller to specify unicode error handling.
(FileSystem.write_text_file): Ditto.
* Scripts/webkitpy/common/system/filesystem_mock.py:
(MockFileSystem.read_text_file): Allow caller to specify unicode error handling.
(MockFileSystem.write_text_file): Ditto.
* Scripts/webkitpy/common/system/filesystem_unittest.py:
(RealFileSystemTest.test_read_text_file_unicode_decode_error): Test reading a file with illegal unicode content.
(RealFileSystemTest.test_write_text_file_unicode_encode_error): Test writing illegal unicode content to a file.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220150 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, skip newly imported WPT HTTPS tests on WK2 only.
cdumez@apple.com [Wed, 2 Aug 2017 19:59:09 +0000 (19:59 +0000)]
Unreviewed, skip newly imported WPT HTTPS tests on WK2 only.

The tests seem to work fine on WK1.

* TestExpectations:
* platform/wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220149 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWe should be OK with the gigacage being disabled on gmalloc
fpizlo@apple.com [Wed, 2 Aug 2017 19:57:50 +0000 (19:57 +0000)]
We should be OK with the gigacage being disabled on gmalloc
https://bugs.webkit.org/show_bug.cgi?id=175082

Reviewed by Michael Saboff.
Source/bmalloc:

This adds Gigacage::shouldBeEnabled(), which returns false when we're using gmalloc or other things
that enable DebugHeap.

* bmalloc/Environment.cpp:
(bmalloc::Environment::Environment):
* bmalloc/Environment.h:
* bmalloc/Gigacage.cpp:
(Gigacage::ensureGigacage):
(Gigacage::shouldBeEnabled):
* bmalloc/Gigacage.h:
* bmalloc/Heap.cpp:
(bmalloc::Heap::Heap):
* bmalloc/Heap.h:

Source/JavaScriptCore:

* jsc.cpp:
(jscmain):

Source/WebKit:

* WebProcess/WebProcess.cpp:
(WebKit::m_webSQLiteDatabaseTracker):

Source/WTF:

* wtf/Gigacage.h:
(Gigacage::shouldBeEnabled):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220148 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Automation: files selected for upload should be checked against values of the...
bburg@apple.com [Wed, 2 Aug 2017 18:35:22 +0000 (18:35 +0000)]
Web Automation: files selected for upload should be checked against values of the 'accept' attribute
https://bugs.webkit.org/show_bug.cgi?id=174803
<rdar://problem/33514190>

Reviewed by Carlos Garcia Campos.

Use the parsed values of the file input element's "accept" attribute to reject
files that don't match the specified values. This is normally done by Safari
using NSOpenPanel, but since a real open panel isn't shown during automation,
it needs to be done here.

Support for limiting accepted files by file extensions will be added when the
same is implemented in the normal code path for the C and Objective-C APIs.

This change is covered by internal WebDriver tests that will be rewritten for the
public Webdriver W3C test suite someday, when safaridriver runs those tests itself.

* UIProcess/Automation/WebAutomationSession.cpp:
(WebKit::WebAutomationSession::handleRunOpenPanel):
Since we already have the proposed files, there's no need to compute a list of
supported extensions based on wildcard MIME types. First check the extension,
then the inferred MIME type for the extension, and then the wildcard MIME type
if the inferred type is not an exact match.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@220147 268f45cc-cd09-0410-ab3c-d52691b4dbfc